joeac/joeac.net-symfony
Archived
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

migrate blog md -> yaml

Browse Source
This commit is contained in:
Joe Carstairs
2025-05-22 08:27:34 +01:00
parent 1538fc7904
commit 8e42c6c3bc
37 changed files with 5247 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

244
scripts/blog-migrated/doctor_who_gayness_church.yaml Normal file
View File

@@ -0,0 +1,244 @@
title: Doctor Who, Gayness, and the Church
description: >-
Series 14 of Doctor Who has a schizophrenic relationship with Christianity.
It’s also gay. I think there might be a connection.
pubDate: 2024-07-08
content: |
I’ve recently finished the most recent series of Doctor Who, series fourteen (or
‘Season One’ as our new benevolent overlords at Disney+ are styling it.) It’s
pretty fun, by the way. I can recommend it for light watching[^1].
There’s a few discussion points coming out of that series that would be worth
dwelling on. But I’ve been particularly thinking about the schizophrenic
attitude the series has taken to Christianity, along with how gay this series
is: and what these two themes might fit together, to give us something important
to say about Jesus Christ, being gay, and the universe.
## The Doctor’s schizophrenic relationship with Christianity
First, let’s look at that schizophrenic attitude to Christianity I mentioned.
The series ticks up an impressive tally of explicit or strongly implicit
references to Christian beliefs and morals, and **none** of them are positive.
In episode 3, _Boom_, the far-future Anglican church has become an army,
with the ranks of the clergy becoming equivalent to arms-bearing ranks in the
soldiery. The Doctor claims that ‘the Church’ has been an army for most of
its history[^2], and that his companion Ruby Sunday (and therefore us) has
been living in a ‘blip’. An all-powerful arms company has tricked the Church
into fighting a non-existent foe in order to keep them buying weapons. The
Doctor attributes their ability to fall for such a deception to religious faith:
> I mean, most armies would notice that they were fighting smoke and shadows,
> but not this lot, Ruby, you know why? They have faith. Faith! The magic word
> that keeps you never having to think for yourself.
In episode 4, _73 Yards_, Kate Lethbridge-Stewart, head of UNIT, makes this
offhand comment, suggesting the oft-repeated claim that all religion is founded
in credulity[^3]:
> That’s what we do, all of us. We see something inexplicable, and invent the
> rules to make it work. Mankind saw the sun rise and created God: or we saw the
> arrival of a Sontaran, one or the other.
In episode 5, _Dot and Bubble_, The Doctor is rejected by the people he is
trying to save. They rant that he is ‘not one of us’, call his claims about the
TARDIS ‘magic’ and ‘voodoo’, and in the same breath assert that it is their
‘God-given duty to maintain the standards of Finetime’ (their space colony).
In episode 6, _Rogue_, a party of murderous aliens turns up at an upper-class
Georgian dance party and demand to be married. The priest denies them, not on
the grounds that they are unrepentant murderers with obviously no intention
of taking their marriage vows seriously, but on the grounds that they are
‘creatures from hell’, _ie_ ‘you don’t look like us’[^4].
There are other references to Christianity here and there which are, in
themselves, neutral or ambiguous. This adds to the sense that Christianity is an
important theme for the series.
But the message is clear. It’s consistently drilled in: Christian beliefs and
morals are stupid and bigoted.
And yet, when the series culminates in its epic two-part finale, the story is
plastered wall-to-wall with Christ-like imagery. It transpires that the Doctor
has unwittingly been acting as an ‘Angel of Death’ by carrying the invisible God
of Death, Sutekh, on his TARDIS, infecting everyone he goes near with the curse
of death. Finally, through one of his children, the curse is unleashed, and
everyone in the universe is given over to death. Even in death, they continue to
suffer, and one of the characters describes themselves as being in ‘hell’. The
Doctor battles Sutekh face-to-face and defeats him. Sutekh becomes one of the
only creatures which The Doctor, usually a staunch pacifist, can bring himself
to kill. Then, the whole world is restored to life. Ruby Sunday even gets to
meet her birth mother for the first time, and their relationship is wonderfully
restored.
So on the one hand, any outward sign of Christianity is despised, but the heart
of the Christian story — the Resurrection — becomes the template for the
climactic redemption story which ties the whole series together.
That’s what I mean when I say this series has a ‘schizophrenic’ relationship
to Christianity.
## What’s this got to do with gayness?
In series fourteen, Doctor Who goes gay.
I’m not playing culture wars here. The Doctor literally kisses a man in episode
six.
But it’s not just that. When we first meet his companion, Ruby Sunday (in the
Christmas special), she’s busy falling in love with a woman. Ncuti Gatwa plays
The Doctor camp (brilliantly by the way). The Doctor refers to Ruby as ‘babes’.
It’s got ‘gay’ written all over it, and this is definitely on purpose.
<figure>
<img
alt="An official promotional picture of Ncuti Gatwa looking lustily into the camera"
src="/images/blog/2024/06/30/ncuti-gatwa-promo-pic.webp"
/>
<figcaption>
Look at this official BBC promo pic and tell me Ncuti Gatwa isn’t playing
The Doctor gay.
</figcaption>
</figure>
From this perspective, it’s not hard to imagine where the hostility to the
Church might come from.
The Church has failed to teach well and Biblically on sexuality, at least in
the last couple of centuries. Everyone has suffered as a result of this, but gay
people often feel the sharpness of this particularly keenly[^5].
The Church has put its weight behind a variety of unhelpful teachings on
sexuality over the centuries[^6]. But two, contradictory, ones stand out as
particularly salient today.
One is **prudishness**. Sex is evil. Sex is the origin of all evil: the Devil
seduced Eve, and Eve seduced Adam, and that’s when it all went downhill. Sex
is naughty. Sex is bad. Remember when Jesus said, ‘whoever looks at a woman to
lust for her has already committed adultery with her in his heart’ (Matt 5:28)?
He was saying that **all sexual desire** is sinful. Remember when Paul advised,
‘It is good for a man not to touch a woman’ (1 Cor 7:1)? He was saying that it’s
**bad** for a man to touch a woman **under any circumstances**.
This is, of course, false and dreadful teaching. The truth is that sex is a gift
from God, given so that we can bear his image by loving each other in this most
intimate and wonderful way. This is the consistent message of both Testaments.
But the idea that there’s somehow something **inherently wrong** with sex has
undeniably been a part of the Church’s teaching since at least the Victorian
period (and possibly a great deal longer than that).
The other is that sex is an **essential aspect of humanity**. If you aren’t
having sex, you’re missing out on an irreplacable part of your created purpose.
This has reacted explosively with the ideas of the Sexual Revolution. The
net result is that we have not so much been **freed to have sex** as we’ve been
**enslaved to have sex**. Virginity is an embarassment – both for men and for
women.
The radical Biblical idea that you can have a completely fulfilled life, deeply
enriched by loving relationships, **without** having sex or getting married,
is forgotten.
Everyone has been harmed by these teachings. But those who experience
significant attraction towards the same sex have been harmed double.
Faced by the impossible demands of bad Church teaching on sexuality on top
of the complexities of living with same-sex desire has left those people with
nowhere to go. They can _just say ‘no’_, and be made to feel that they’re
missing out on completing their full humanity. Or they can indulge their
same-sex attraction, and be judged not only promiscuous but a pervert to boot.
No wonder so many gay people have given up on the Church (and I haven’t even
talked about discrimination or violence towards gay people).
And no wonder that The Doctor has become strikingly anti-Christian at the same
time as it has become strikingly gay.
## What next?
Doctor Who’s criticisms of the Church hurt when they hit the mark. They hurt
more when they’re unfair. But that’s not the point. That point is this: the
Church has lost control of the conversation. And we’ve lost control of the
conversation because we threw it away. We threw it away encased in bomb-proof
concrete and left it to sink to the bottom of the deepest available ocean
trench.
If we in the Church are feeling hurt, we should start by feeling hurt by
our own sin.
The work of regaining trust on the question of sexuality will be the work of
decades: and that’s if we start working full-pelt right now. But there is hope.
I was encouraged last December by the visit to my local church, Bruntsfield
Evangelical, of _Living Out_[^7], a charity dedicated to helping churches across
the UK talk about sexuality. Ed and Andy, both same-sex attracted men, led us
through talks and discussions, and played recordings of perspectives from their
same-sex attracted female colleagues who couldn’t make it on the day.
They were primarily there to talk to us about how to support people like them:
same-sex attracted Christians who might be in our church who believe they are
called to singleness, helping them to thrive, living their true selves openly,
surrounded by love and empowered to share their love with others, just as we
want for everyone in our church. In fact, I heard some of the best news I’d ever
heard about my own sexuality, even though I am in the minority of people who
have never experienced significant same-sex attraction.
Whether or not you agree with their stance that the right place for sex is
within a marriage between a man and a woman, charities like _Living Out_ are
driving the Church in the right direction: serious, Biblical sexual doctrine
which helps us to realise in practice how we are all made in the image of the
God of love.
And how needed! How desperately needed! And Doctor Who gives us a little glimpse
of that, too.
Because not only is the series apparently anti-Christian, it also expresses a
need for Christian salvation. The Resurrection story is one which everyone needs
to hear, to have their death transformed to life, to have their full humanity
affirmed and celebrated and tended and to delight in it and to see it flourish.
To know the God of love and life, behind all and over all, with a plan and a
means to defeat the grip of death on the world.
So let’s keep having those frank conversations about sexuality in the Church.
It matters for all of us.
[^1]: Unless you’re bothered by
[plotholes big enough to fly a TARDIS through](https://www.doctorwhotv.co.uk/doctor-who-opinion-empire-death-2-102716.htm).
Oh, and you can probably skip episodes 1 and 3.
[^2]: For the record, this is flatly false. Though exceptions are widespread, the
overwhelming mainstream opinion is that priests should not bear arms. This is
not new, but has been the consistent orthodoxy, taken straight from Jesus’
teachings by the Church Fathers and maintained constantly since then. Can the
Church be legitimately criticised for its use of violence? Absolutely, let’s
have that discussion. Has the Church almost always been a **literal army**? No.
The Church has **never** been an army.
[^3]: See _God is Not Great_ by Christopher Hitchens, Chapter 11 for an epitome
example of this put forward by a prominent atheist apologist.
[^4]: In the spirit of charity, I have to accept that this admits of multiple
interpretations. It is possible that by calling them ‘creatures from hell’,
the priest is referring precisely to their unrepentant murderousness. However,
the priest has already tried to avoid the creature’s gaze by the pitiably
schoolboyish ploy of **looking elsewhere**, priming the viewers to think
that the priest is a silly coward: certainly not the kind to make a noble,
principled defence of justice with his head in the jaws of death. Plus, if
we interpret ‘creatures from hell’ as meaning ‘you don’t look like us so you
must be evil’, that would fit with the consistent framing of the series: that
religious morality is equivalent to bigotry.
[^5]: Though it must be remembered that many gay people have remained and
thrived within the Church in spite of the challenges, serving in diverse ways,
not least the priesthood.
[^6]: I found the Ezra Institute to give a good
[introduction to the history of Church teachings on sexuality](https://www.ezrainstitute.com/resource-library/articles/sex-and-the-history-of-christianity).
[^7]: For a great and humane introduction to what these guys are like, have a
listen to this [Living Out podcast episode](https://www.livingout.org/resources/podcasts/68/misstep-7-godliness-is-heterosexuality-the-plausibility-problem-7).
I get that if you haven’t grappled with issues like these before, it can be
really challenging, and you might find it hard to trust people who take a
different view to yourself. I know: I’ve been there myself! The best way to
start is to listen to a human voice.

200
scripts/blog-migrated/does_resurrection_ground_works.yaml Normal file
View File

@@ -0,0 +1,200 @@
title: Does resurrection doctrine give us unique reasons to work for justice?
description: >-
Tom Wright claims that the only Christian grounds for striving for justice now
is in resurrection doctrine. I'm not convinced.
pubDate: 2025-05-04
content: |
I've been reading Tom Wright's _Surprised By Hope_, defending his orthodox view
on resurrection. One of his key claims is that only by accepting the orthodox
position on resurrection can Christians justify striving for justice on earth.
To argue this, he needs to first show that resurrection doctrine does justify
striving for justice on earth, and secondly that the available alternatives fail
to do so. Firstly, the positive argument.
## Does resurrection give us reasons to work for justice?
Wright's argument depends on his view on what God's ultimate future will look
like: the present creation will not be abandoned, destroyed, or replaced, but
physically transformed into the new creation.
He argues that our work now has value, because, at the time when God transforms
the old world into the new, he will incorporate the outcomes of our good works
into the new creation, like an architect incorporating the works of many
stonemasons into a great cathedral.
He argues this on the basis of 1 Cor 15. I found it a struggle to find a good
justification for Wright's view in chapter 15 alone, but I did find some crucial
context in chapter 3, just before he begins the first of his many exhortations
to the Corinthians. It would be best to read the whole chapter, but here is
verses 11-15:
> For no one can lay any foundation other than the one already laid, which is
> Jesus Christ. If anyone builds on this foundation using gold, silver, costly
> stones, wood, hay or straw, their work will be shown for what it is, because
> the Day will bring it to light. It will be revealed with fire, and the fire
> will test the quality of each person’s work. If what has been built survives,
> the builder will receive a reward. If it is burned up, the builder will suffer
> loss but yet will be saved—even though only as one escaping through the
> flames.
This at least implies that the outcome of all our work will persist at least
until the time of judgement, when it will undergo testing, and those that pass
the test will enjoy a reward. You could read this as being like a quality check,
with God dishing out benefits to those that pass his assessment. But you could
read it in another way, more favourable to Wright. The works will be proven, and
the ones that withstand the process will themselves generate a benefit. You
could think of it like baking: when you put a cake in the oven, you prove
whether or not you got the recipe right; if not, it goes in the bin, and if you
did get it right, you get to enjoy the cake. So I agree that 1 Cor provides a
reason to think that the outcome of our works will somehow persist until the
time of judgement.
However, Wright doesn't just think that our works will persist until the time of
judgement, he also believes that they will at that time be transformed and then
incorporated into the new creation. 1 Cor doesn't directly justify this view. It
is, at least, coherent. Baking a cake in the oven transforms the dough.
So it seems reasonable to me to use 1 Cor to justify Wright's view that our
works will be transformed into the new Creation. However, that's before
considering any counter-arguments, and I have two which concern me.
One is that, as Ecclesiastes points out, the profits of our work will be laid to
waste by time. We don't know when God is going to bring about the new creation.
It could be tomorrow, and we should be behaving today in light of that
possibility - but of all the available possibilities, most of them are in the
distant future, so we should expect on average that there will be a long
interval between my deeds today and the judgement of them. Since time lays waste
to all our endeavours, we should expect that the profits of our work will have
vanished long before the judgement. Therefore, it would be unreasonable to work
for justice in the hope that our work will persist until the time of judgement
in order to be incorporated into the new creation, as Wright argues we should
do.
This first problem is probably the most important, but it's vulnerable to some
counter-arguments, which rather miss the point. So I'm going to move on to my
second problem, which is more niche, but which I think holds stronger against
counter-arguments.
In 1845, the HMS Erebus and HMS Terror left Britain on an expedition to explore
the Northwest Passage. In July 1845, they were spotted by whalers in Baffin Bay.
None of the sailors were ever seen alive again by Europeans. Both ships became
locked into ice in September 1846, and probably never sailed again. In April
1848, the remaining crew began a desperate 250-mile march to the nearest
European outpost. It is almost certain that all of them died.
Now, suppose that one of those sailors, between 1846 and 1848, did something
good. Maybe a major act of heroism, or maybe a small, kind word. I think it's
reasonable to suppose that someone, at some time, did something good.
All the profits of that good act have now completely perished. They are not
recorded in the small records the sailors left behind. None of the sailors who
benefited lived to pass on the benefits. None survived to pass on the light of
justice to the next generation.
At the time of judgement, then, there will be no remnant of this good act for
God to test, transform and incorporate into his new creation.
Does this mean that the good act was wasted? Was it only worth a shot just in
case, despite all the odds, the sailors made it home again?
I think this is a cruel conclusion: but it seems to be where Wright must go.
If it is implausible that the profits of all our good works will make it through
to the time of judgement, then we have to accept that 1 Cor 3 is true only in
some metaphorical sense, not that our works will literally persist in order to
be tested. And if that is true, then the premise of Wright's argument is false:
our work will not necessarily be incorporated into the new creation, so that
cannot generate reasons for working for justice now.
I am not convinced by Wright's positive argument. I would like to consider
alternative interpretations of 1 Cor. This is what Wright goes on to do.
## Do alternative views give us reason to work for justice?
The first view which Wright considers is the gnostic view that resurrection is
just an afterlife in heaven. He argues that this does not generate reasons for
working for justice now, but his argument is really the converse of the argument
in favour of his own view, which I've already considered above, so I'll move on.
The other alternative Wright considers is what he calls 'evolutionary optimism'.
You might also call this 'progressivism'. By this he means the view that the new
creation will be made the Church gradually building upon its own works,
generation after generation, approaching and eventually achieving God's perfect
standard by its works. This is the Victorian optimism which is still a powerful
force in our politics: that history is building upon itself, and progressing
from barbarism to civilisation, from brutishness to beauty, from tragedy to
justice.
He argues that this view, too, does not give us reasons to work for justice now.
His argument is intriguing: if our work for justice is condemned to only ever be
partially successful, then we have no reason to do it. In fact, contra the
evolutionary optimist, no amount of hard work on our part will ever achieve
perfect justice, and therefore if bringing in the new creation is all about our
works, we have no reason to strive for justice.
I'm intrigued by the premise that if we know that our work will at best be
partially successful, then we have no reason to do it.
This isn't how we ordinarily think: typically, if I think attempting to go to
the gym twice a week is going to be partially successful, I would say that this
generates a reason for me to go to the gym.
But I sense there may be an interesting meta-ethical thesis here: perhaps what
we ordinarily call 'partial success' is in fact a mistake, papering over what is
in fact simply a failure.
There's an obvious error theory: we had to create the concept of the 'partial
success' in order to generate reasons in the world as it appears, the world as
described in Ecclesiastes as 'vanity', where the best-laid plans of mice and men
gang aft agley, and all our works crumble into the dust eventually. In order to
think we had reasons at all, we needed to invent the concept of the 'partial
success'.
But is there a good argument for the view that there is no such thing as a
partial success?
In 1915, the HMS Endurance, under the command of Captain Ernest Shackleton,
became stuck fast in Antarctic pack ice in the Weddell Sea. In the face of
extraordinary challenges, Shackleton vowed to bring all his sailors back to the
UK alive: and in one of the most famous exploits of Antarctic exploration, he
succeeded.
It would have been wrong, had Shackleton vowed only to bring back _most_ of his
sailors. We know that he would have been able to bring _all_ of them back,
because he in fact did so. To strive for less would have been negligent.
This is to illustrate a general principle: we ought to strive for the best that
we are able.
Grant that necessarily, no particular justice is inevitable. It follows that
necessarily, it is possible to prevent all injustice. Therefore, perfect justice
is achievable.
And yet we _know_ that we will not achieve perfect justice. It's way too hard.
It does seem that I've proven a contradiction: both that perfect justice is
possible and that it is impossible. I expect these are two different kinds of
modality. I'm not too bothered to carefully distinguish them, as long we agree
that these two things can both be true in some sense. By analogy, consider that,
if you can run a mile in so many seconds, you can run it in a second less; that,
by sorites, it follows that you can run a mile in a minute; and that you
obviously cannot run a mile in a minute: it is too hard. Perfect justice is
perhaps a little like this: it is achievable in the sense that it is physically
possible for us to achieve it, but unachievable in the sense that it's way too
hard.
So, since perfect justice is in some sense achievable, it follows that we ought
to strive for it. But since we know we will not achieve perfect justice, it
follows that we cannot have a reason to strive for it: we cannot genuinely
strive for what we know we cannot do. Therefore, if you reject that God will
work to transform our world of vanity into something fundamentally different
where perfect justice is not only genuinely achievable but actually realised,
then you will be stuck in this hopeless tension, where you both must bring about
perfect justice, and have no reason to do it, because you have no hope of
success.
Where from here? I would really like to find alternative interpretations of 1 Cor,
and weigh them up against Wright's interpretation. It may be that, whatever the
counter-arguments, Wright's view is the strongest available. It may not.

94
scripts/blog-migrated/easter.yaml Normal file
View File

@@ -0,0 +1,94 @@
title: Why Easter is the best week of the year
description: >-
Based on a talk given to my colleagues at
<a href="https://www.scottlogic.co.uk">Scott Logic</a> for Maundy
Thursday, 2024.
pubDate: 2024-03-30
content: |
As you might have noticed, it is Easter this week! So I'd like to take five or
five minutes of your time to share why I – and about two billion other humans
going about the place just now – think Easter is the best week of the year. And
it's got something to do with a special Christian ritual called Communion.
Communion, at its heart, is about as simple a ritual as you can get. You get
together with a bunch of other people. You share some bread, and you share some
wine.
And it’s because of this ritual that so many people regard Easter as the best
week of the year. I want to explain to you why that is, and more than that, I
want to convince you that Easter is the best week of the year for you, too!
If you’ve passed by _The Hub_ at the top of Johnstone Terrace here in
Edinburgh recently, you might have notice the banner which is draped over
the railings just now – reading, ‘RITUALS THAT UNITE US.’
Now, that might seem like an odd idea. But wouldn’t that be great, if we
actually had a ritual which could unite us? Because the world could surely do
with a bit more unity right now. The world seems so divided, and sometimes it
seems like there’s no hope for real unity.
We can see that in our politics. We’re divided about foreign policy, about
taxation policy, about trade policy, about environmental policy.
And the conflicts that we have in this country seem pretty trivial when we
remember the conflicts that are playing out in other parts of the world right
now. In Israel and Gaza. In Sudan. In Russia and Ukraine.
And there’s plenty of conflict happening on the small scale, too. Often it’s the
smallest-scale conflicts which hurt us the most deeply. Your landlord pushes you
around. That friend you trusted like no-one else in the world lets you down. The
partner or spouse you loved like no-one else in the world – you end up fighting.
It’s possible that you’re going to be reading this right now with a heavy heart
because of a broken relationship in their life. And doesn’t that hurt more than
anything else we know?
When the world is groaning so heavily under the weight of conflict, and some
banner on _The Hub_ tells us a ritual can unite us, that seems so out of
proportion to the scale of the problem, doesn’t it? What can a ritual do? A bit
of old superstition? An excuse to divide people, maybe – what can a ritual do
to unite us?
Well, two thousand years ago, a man had a meal with his friends. Together, they
shared a meal of bread and wine – which, in that time and place, was the most
ordinary meal imaginable.
And yet, in that most ordinary event imaginable, something was happening which
was totally unimaginable. As this man, Jesus, shared the elements of this meal,
he made some extraordinary statements about what he was doing: ‘take, eat, this
is my body’ – ‘drink this, all of you; this is my blood of the new covenant.’
(The word ‘covenant’ means a promise.)
He told them he wasn’t just giving them bread and wine, he was giving his body
and his blood, and a promise.
Before Jesus ate another meal, he was flogged and nailed to a cross. His blood
was spilt and his body broken, even to death.
And yet, that wasn’t the end of the Easter story. Because three days later,
mourners turned up at Jesus’ tomb to pay their respects, and found the tomb
empty, the stone rolled away. Then they became the first of crowds of
incredulous eyewitnesses to see Jesus, the same Jesus who was killed on a cross,
alive.
Some magic trick, right? But this matters a hell of a lot more than just some
magic trick. Because Jesus became the first person in history to prove that you
really can both have your cake and eat it. He gave his life, and lived! As a
result, we can have his life and our own. We can join with Jesus through the
ritual of Communion which he established, and thereby, through Jesus’ body, join
together with everyone else who takes part in that ritual, as one body. Then we
can start living our brand-new, full-fat, original-recipe life overflowing with
generosity where we too can both give our life to others and enjoy it ourselves.
Indeed, Jesus taught us and showed us that it’s precisely by giving our lives to
others that we get to truly live ourselves.
This is why, in spite of all the division which persists in the world today, two
billion people regard this week as the best week of the year. Two billion
people, from every nation on Earth, speaking thousands of languages, of every
age and culture and gender and race, who defy the divisions of this world to
insist on joining together as one body in Jesus.
Because Jesus’ new covenant, his promise to all of us, is that in an apparently
hopelessly divided world, there exists real hope for unity. And that’s why
Easter is the best week of the year.

258
scripts/blog-migrated/euhwc_toast_to_the_lasses_2024.yaml Normal file
View File

@@ -0,0 +1,258 @@
title: EUHWC Toast to the Lassies 2024
hidden: true
description: >-
At the <a href="https://www.euhwc.co.uk">EUHWC</a> Burns meet in
Ullapool last weekend, I had the last privilege of giving the Toast to
the Lassies. Particularly for the benefit of those who weren’t there,
here it is in full!
pubDate: 2024-01-29
content: |
Had Burns, instead of his sweet bonnie Jean,<br>
his skills poetical for to mature<br>
had any one of our club’s lassies seen<br>
he would forever have remained obscure.<br>
If he had nothing but this box of worms<br>
Scotia would have been poorer, that I’m sure.<br>
Now none of us can claim to be a Burns,<br>
I’m no poetic master, still, I’ll have a punt,<br>
though let’s be clear, I’ll do it on my terms.<br>
I’ve everywhere avoided being blunt -<br>
politeness matters more than any schema -<br>
but it is hard when Isla’s such a cunt.<br>
It was a challenge to produce a terza rima<br>
I could recite withouten snoring;<br>
you’ve been so stiff I thought youse had oedema.<br>
The bother is this year is you’ll all been boring:<br>
no drugs, no sex, no gossiping or lies,<br>
no rock and roll, and hardly any whoring.<br>
But hey well, rules is rules, I’ve had to try!<br>
At least it can’t be worse than the reply.<br>
I’ll start with Audrey, the club’s senior member,<br>
for if there’s something that I say which disconcerts her,<br>
it’s fine: the poor old girl, she won’t remember.<br>
She likes to let us think she’s a hard worker<br>
but we’re electing a third social sec…<br>
it’s pretty clear she’s just another shirker.<br>
This lady, half American, half Czech,<br>
for study, moved to Scotland for to do<br>
American history – really, what the heck?<br>
The club is so much louder thanks to you:<br>
impressive vocals for just five foot two.<br>
That woman, Willow, reggles is bespeckled<br>
with her sickle and her fishing tackle<br>
shackled by the shins while she is heckled;<br>
the way that Willow waddles maks me cackle<br>
like a speckled jackal getting tickles,<br>
worth a shekel in the tabernacle;<br>
I chuckle muckle at her love of pickles<br>
which she wiggles when she has the heart<br>
while work for the Committee’s fickle trickles.<br>
Her modus operandi: <em>you can’t rush art.</em><br>
Her reimbursements programme’s going great;<br>
any day now, she’ll maybe even start.<br>
She cannot walk without Audrey, her mate:<br>
I wonder when they’re going to consummate.<br>
Although they make them pretty tough in Peebles,<br>
the thought of actually going up a peak<br>
fills Shona Lewis with the heeble-jeebles.<br>
New car? We miss your beautiful antique!<br>
How long before this one’s also up a creek?<br>
Once there was a lass called Hannah Collier<br>
whom even hell below regarded nasty,<br>
deeply despised by all that dwells there.<br>
Dating’s proceeding slowly for our lassie;<br>
not far from giving up til she beguiles<br>
a hot Italian in Southsider: classy!<br>
At first, Michaelo seems to be all smiles<br>
till it transpires he’s one of Dante’s demons…<br>
I guess it’s back to posters of Harry Styles.<br>
One day you’ll get a decent boy, keep dreamin;<br>
somewhere there waits a handsome Mr Collier.<br>
Hopefully when she meets him she’ll no be steamin.<br>
Hannah, I’m not sure why you chose to maul your<br>
poor skeleton at Subway (she’s still tetchy)<br>
and then abandon what remains of all your<br>
dignity at Ryvoan with a Frenchie!<br>
I think he wishes that he never met ye.<br>
And has a quiet Felicia e’er been seen?<br>
The energy she has is frankly wild.<br>
I’ve never seen a hillwalker so keen!<br>
<i lang="de">Ssie ischt raschtlos und nie gelangweilt</i>.<br>
She eats raw oats with soggy protein powder:<br>
a camping pot has ne’er been worse defiled.<br>
She uses what her Maker has endowed her<br>
with: her recorder skills are off the charts;<br>
youse think I’m joking, but I wouldn’t doubt her!<br>
This lass of the land of the Rot-Gold-Schwarz<br>
will soon depart, though long we might beseech you<br>
to stay. Of course, you’ll break all of our hearts,<br>
but mine most of all. Any time, Felicia,<br>
Creag Meagaidh calls, I know routes up the rear<br>
dark and under-explored that I can teach you!<br>
I won’t deny I think it’s rather queer<br>
the things you do with chickpeas, but no matter.<br>
You’re keen, you’re quick, you’re cool, that much is clear.<br>
In fact, I think you’d make a damn good faffer:<br>
swoop down on distilleries like the Luftwaffe.<br>
And now we come to our girl Emily Topness!<br>
You’re keen for social sec. You’d suit the role<br>
because… I’m not quite sure, it’s embdy’s guess.<br>
We met your sister, and she was just as dull.<br>
No, please drone on about Icelandic soil!<br>
Poor Joe here down the front’s bored out his skull.<br>
And since I mentioned Joe – I hate to spoil<br>
it for you – but you’ve got the inferior Joe,<br>
by Jove, no joke, it’s Jock here’s got the style!<br>
Nah, write the boy a sonnet, get in the flow,<br>
Whatever you produce’ll beat by thrice<br>
your Masters thesis. What’d you got to show<br>
for months of hunting for the butterflies?<br>
‘There weren’t any.’ Oh, and have some sense,<br>
cos I’ve heard rumours – I assume they’re lies –<br>
you’ve called yourself the ‘poet in residence.’<br>
You know you can’t compete, drop the pretence.<br>
Tereza was our gear sec for last year.<br>
She helped herself to stuff: that’s factual.<br>
Now when she asks to loan a bit club gear<br>
we have to ask her to provide collateral.<br>
She picked up tin whistle pretty sharp!<br>
Which is to say, she’s not a natural.<br>
She’s nowhere happier than under tarp<br>
gazing up at the moon and stars alone<br>
somewhere distant and remote like Glen Tarff.<br>
Now what to say about Lucy Ma-the-soooon....<br>
she likes… to faff… mm hmmm… well, moving on!<br>
And now we come to Emilie the French.<br>
She seems to be nice on the trips we see her<br>
but my distrust of frogs will ne’er be quenched.<br>
Claims she’s a ‘pharmacist’? So she’s a dealer.<br>
Need some pills in a pinch? You call, she’s there<br>
at your door in her rally-approved four-wheeler.<br>
One question we have is, why are you here?<br>
Most folk are in uni, you’ve no refutin<br>
you were kicked out after second year!<br>
Now the Engineering grad, Sophia Newton.<br>
Your namesake, Isaac, was a man convicted,<br>
constructed calculus; but no computin,<br>
not even Isaac’s, could’ve e’er predicted<br>
you’d drop the Eng for creative writing!<br>
now that’s what I would call a self-inflicted<br>
inflection point! It must be quite enlightening,<br>
but that doesn’t excuse when you give us an earful.<br>
The blood boils in our veins, the rage heightening,<br>
and you’re an American, that makes me fearful.<br>
What’s your secret? You have us knackered!<br>
What are you on to always be so cheerful?<br>
Now we approach the topic of Merzbacher.<br>
Wait, she’s not here? Abandoned ship?<br>
She says she’s informatics: so she’s a hacker?<br>
She has strong views, she lets her anger rip.<br>
Poor George got an earful, full of future advice,<br>
but why hasn’t she been on another club trip?<br>
We’re cruel to focus on this list of vice;<br>
the fact remains: she’s headstrong and nice.<br>
On Skye, a lady gave her poles to Sasha,<br>
which was really nice - I mean just the best -<br>
but Sasha really didn’t have to flash her.<br>
Quick history lesson: way back, RBS<br>
led the banking system to self-destruct<br>
and left taxpayers to pick up the mess.<br>
Since then, the name’s so irredeemably fucked<br>
they’ve had to ditch the brand once and for all.<br>
There’s one lassie who I need not instruct<br>
What, these days, the Royal Bank is called<br>
cos NatWest’s nasty history of scandal<br>
didn’t stop Booth from working there at all.<br>
Nothing motivates her more than to trample<br>
upon the working class. They set her free.<br>
She sank the pound quicker than the Belgrano,<br>
because ‘there is no such thing as society,’<br>
that’s how it is, is it? All right, I see.<br>
Now, coming all the way from Glenmore Lodge,<br>
it’s Ellie’s turn! We have done what we can,<br>
although I’m scared what she’ll put in my squash.<br>
She wasn’t into Benji, but listen man,<br>
you’re lucky that you dodged her drunken benders.<br>
You’ll wake up in a tent in Kyrgystan,<br>
as for how you got there, no-one remembers,<br>
and if you’d known you’d be sleeping next to Ellie,<br>
you would’ve brought some fucking ear defenders.<br>
She’ll wrap you in bubblewrap, from your ears to your belly,<br>
cotton clothes for none, and no complaining,<br>
applying safety to the max, spare socks in your wellies.<br>
She’s always at her Mountain Leader training,<br>
practicing her night nav in the locale,<br>
pursuing QMDs - unless it’s raining.<br>
But some water should not scare our gal!<br>
She’s had much experience with the wet as of late:<br>
after all, she got on well with our navy pal.<br>
What was the age of that particular first mate?<br>
Older than your ex - always part of the plan?<br>
Ah, of course! He was a spry twenty-eight!<br>
Youth’s for the losers, let’s get you a real man,<br>
mature and rugged, but kind and astute?<br>
Just make sure he’s not as old as your gran.<br>
One request we all have is you ditch the uke:<br>
never have strings been pluckèd quite so shitely;<br>
we would all much rather be hit by a nuke.<br>
And please shut up about your nice society.<br>
We are all glad you had a fun summer,<br>
but bringing it up throws us right back to sobriety.<br>
To lose you of course would be a bummer:<br>
that is, for your carefully groomed newcomers.<br>
Now time for the main woman, El Presidente!<br>
To here, it’s been like getting stones to bleed,<br>
but in Isla Burslem’s case we’ve material aplenty!<br>
As Holy Scripture says, ‘let those who lead<br>
well be worthy of double honour,’ so<br>
your bit is double length – it’s quite the screed!<br>
I’ll start off with her brilliant boyfriend – oh!<br>
Not boyfriend! Friend? To me this rather smacks<br>
of low commitment, but what do I know?<br>
So far, he’s disappointing, but on track.<br>
What’s he up to Isla: seven minutes? neat!<br>
Despite that, he is never holding back<br>
your blossoming romance with Dr Peat.<br>
Don’t deny it, that launch was pretty hard!<br>
It’s fifth base next: that’s photos of his feet.<br>
It’s fair to say her reputation’s marred.<br>
We all regret that we did once anoint<br>
her President: her premiership’s ill-starred.<br>
Hey - you’re meant to be in charge of this joint!<br>
You’re seldom seen cos of the mountaineering<br>
meets that you’re always on. You’d made your point<br>
before you chose to go off disappearing<br>
to <em>New Zealand</em>… we get the message! Plus<br>
we’ve had enough of all your domineering:<br>
maybe it’s time we put you on a bus!<br>
Nah, I’m just joking. All I’ve said’s refutable.<br>
But the boys, we mean this next bit, all of us,<br>
so stop me Isla if this isn’t suitable<br>
but honestly we think your mum is beautiful.<br>
Alas, I have to bring an end to this rhyme.<br>
I know it wasn’t much, in our defence,<br>
the fact you used ChatGPT’s a crime.<br>
I hope I’ve not caused over much offence<br>
don’t worry, that is it, I’ve said my bit,<br>
so I’ll turn from the ladies to the gents.<br>
Yeah, don’t look away now, we wrote this shit!<br>
I see you looking at your laces, Chris!<br>
Wit without real goodwill is not legit,<br>
so boys, don’t send sincerity to piss!<br>
Why did God say he’d take our hearts of stone<br>
and give us hearts of flesh? For this, for this!<br>
Here is flesh of our flesh, bone of our bone;<br>
love, and love nothing more but God alone.<br>

249
scripts/blog-migrated/god_is_not_great.yaml Normal file
View File

@@ -0,0 +1,249 @@
title: God Is Not Great, initial thoughts
description: >-
My spark notes on Hitchen's classic 2007 polemic against religion, plus some
initial thoughts on how I want to respond to it.
pubDate: 2024-04-14
content: |
These are my 'spark notes' on _God Is Not Great_, Christopher Hitchen's classic
2007 polemic against religion in all its forms, and call to adopt secular
humanism as its rightful replacement.
The book can be coherently read as a collection of independent essays. That
said, with a few exceptions, each chapter in _God Is Not Great_ contributes to
one of three main themes, and I think this is a helpful way of summarising the
overall movement of the book. The three themes I've identified are as follows:
<ol>
<li>
<a href="#1-religion-is-evil">Religion is evil</a>
</li>
<li>
<a href="#2-religion-is-untrue">Religion is untrue</a>
</li>
<li>
<a href="#3-secular-humanism-is-a-better-alternative-to-religion">Secular humanism is a better alternative to religion</a>
</li>
</ol>
As for the exceptions: Chapters 13 and 17 in part contribute to both the first
theme and the third, while Chapters 1, 3, 12 and 14 don't fit into these broad
themes, and are self-standing.
## 1. Religion is evil
- Chp 2
- Religion is violent, because:
- It has to be missionary, because it is insecure in its own beliefs (p17)
- It provokes tribalistic conflict, in a similar manner to racism (pp35-36)
- Chp 4
- Religion is bad for your health, because:
- Faith in medicine is a threat to religion's thrall (p47)
- Religious doctrines may contradict sound medical advice
- The right to religious freedom may be abused to deflect criticism of
unhealthy practices (p50)
- Religion has a special relationship with child abuse, and is incapable of
accepting open moral criticism of itself for this
- Chp 13
- Religion tends towards evil, because it requires fanaticism in order to
spread, and fanaticism tends towards evil (p192)
- Chp 16
- Religion causes child abuse, because:
- Terrifying children with eschatology is child abuse
- Religious education is propaganda and should not be inflicted on children
who are not yet mature enough to respond to it rationally
- Religion consistently mandates cruel genital mutilation of children
- Christians and Muslims have spread misinformation about masturbation,
which leads to dangerously sexually repressed adult men, which in turn
leads to sexual abuse of children by those men (as well as sexual abuse of
women)
- Child abuse in churches is not a case of a few bad eggs, it is
institutional and based on an ideological need to control the minds and
sexual organs of children
- Religion institutionalised torture in medieval Europe
- Religion makes honest and nuanced debate about abortion impossible, because:
- Nuanced debate is pushed out by extreme and implausible religious doctrine
- Religious people would rather use the unborn as objects of doctrine than
human beings in need of protection
- Chp 17
- Religion is the only reason anti-Semitism is possible
- Religion in its fullest expression is indistinguishable from
totalitarianism, because:
- The defining characteristic of both religion and totalitarianism is the
absolute right to rule of the ruler, even when they rule with caprice
- Religion and totalitarianism are also characterised by the need to
extinguish heresy with violence
- Religion and totalitarianism alike must propose a total solution to all
life's problems, require blind faith from its adherents, and demand all
aspects of life public and private be submitted to total supervision. This
doesn't bring out the best in us
- Religious/totalitarian systems are unable to take accountability and
therefore improve over time, in contrast to secular humanist systems
- History has proven this as fact:
- In the ancient world, religious totalitarianism was the normal form of
government
- Calvin, the inspiration for the Presbyterian Christian tradition, which
included South African apartheid, was the epitome of a totalitarian
dictator, demanding total control on the private lives of his citizens
in Geneva, on the pain of humiliation in this life and eternal torment
in the next
- History has also proven that, rather than standing in opposition to the
supposedly secular totalitarianisms of the twentieth century, religion
actually aided and abetted totalitarianism:
- Rome supported fascist movements throughout Europe, including Italy,
Hungary, Spain, and Ireland
- Rome accommodated Naziism by handing over control of its schools,
permitting the use of parish records to identify those with Jewish
ancestry, disbanding Catholic opposition political parties, declaring
Hitler's birthday a Church holiday, and running the 'rat line' to South
America after the military defeat of Naziism
- Although not quite as arse-licking as the Vatican, Germany's Protestant
churches also mostly capitulated to Nazi totalitarianism
- Japanese soldiers committed enormous atrocities across the Far East in
the name of their god-emperor, Hirohito
- The strategy of the Communists was first, to use religion as a prop to
gain power, and then to replace religion with itself. Notice the
striking commonalities between religion and communist totalitarianism:
- Infallible leaders
- A permanent war on heresy
- Institutionalised torture
- Scapegoating the innocent rather than accepting accountability for
failures
- Justifying any means necessary in order to achieve an ultimate end
## 2. Religion is untrue
- Chp 5
- Religion was a barbaric attempt to explain physical phenomena. Science now
does a much better job, so religion can be discarded as a redundant theory
- Chp 6
- Religion is 'solipsistic', which is to say:
- Religion divides the world into an in-group and an out-group
- When the in-group receives good fortune, that is interpreted as God's
blessing
- When the in-group receives bad fortune, this is inexplicable
- Whatever happens to the out-group is irrelevant
- This is a redundant theory which explains little, and therefore we should
not believe it
- Chp 7
- We should believe that the Pentateuch is a fiction, because:
- Exodus is inconsistent with the archaeological evidence
- Textual evidence in Deuteronomy suggests the texts were written much later
than the supposed events were supposed to have taken place
- Throughout the Pentateuch, Moses is referred to in the third person, which
is not consistent with the claim that Moses himself authored it
- The Pentateuch contains events we should _hope_ to be false, such as Moses
ordering multiple massacres, and the Ten Commandments classifying wives as
their husbands' property
- The Pentateuch - indeed, the whole Bible - is limited in scope to a small
corner of the Middle East, which is not consistent with the universal
nature of the supposed God who is supposed to have inspired its authorship
- Chp 8
- We should not believe the Gospels, because:
- Matthew and Luke disagree on the virgin conception
- Matthew and Luke disagree on the genealogy of Jesus
- Matthew and Luke disagree on when Mary and Jesus escaped to Egypt
- Luke dates the birth of Christ during both the reign of Herod in Judea
and the governorship of Quirinius in Syria, but these two events did not
overlap, so this is impossible
- As far as we know, the Romans did not, and would not, demand that the
people assemble in one place in order to be counted for a census
- The Gospels disagree about the Sermon on the Mount
- The Gospels disagree about the Anointing of Jesus
- The Gospels disagree about the Crucifixion and Resurrection of Jesus
- The Gospels disagree about the betrayal of Judas
- The Gospels disagree about the denial of Peter
- John suggests he thinks Jesus was probably born in Galilee to an
ordinary family with no proven link to King David's genealogy
- Mary's attested behaviour during the ministry of Jesus is not consistent
with the Nativity story
- John 8:3-11 (the stoning of the woman caught in adultery) is a later
insertion to John's Gospel
- Chp 9: the Koran is borrowed from a hotchpotch of Jewish and Christian
myths
- Chp 10
- Miracles (such as the Resurrection of Jesus) should not be believed,
because:
- See Hume's _On Miracles_
- Extraordinary claims require extraordinary evidence
- Believing miracles is comparable to believing reports of aliens (p144)
- Chp 11
- Religions are founded in credulity
- Chp 15
- The crucifixion story makes no sense, because:
- I'm supposed to be morally responsible for Adam's sin, but I amn't!
- Even Adam can't be fully blamed for Adam's sin, because he was set up!
- The Jews who crucified Jesus are supposed to be blameworthy for their
crime, even though the crucifixion was allegedly both necessary and
inevitable
- I'm allegedly given the free will to either accept or deny the offer,
even though denying the offer will lead to an eternity of torment: any
sensible God who cared a mite would not have given me the choice
- Religious rules are impossible to follow, and this leads to either
spiritual policing, organised hypocrisy, or both
## 3. Secular humanism is a better alternative to religion
- Chp 13
- Faith has inspired great heroism. But the heroism is better explained by the
heroes' humanism than it is by their faith itself
- For example, Martin Luther King Junior didn't really preach Christianity,
because he preached forgiveness, while Christ preached eternal torment for
the 'inattentive' (pp175-6)
- Chp 17
- Religion is the only thing sustaining anti-Semitism
- Secular humanist political systems can take accountability, respond to
criticism and improve over time, while religious systems exempt themselves
from legitimate criticism, stifling progress
- The only alternative to totalitarianism is pluralism, which is inherently
secular
- Chp 18
- Secular humanism has been a powerful positive force throughout history in
face of oppression by religion, as proved by example:
- Socrates proved that conscience is innate, and that a great way to mock
dogma is with satire which pretends to accept that dogma
- Lucretius, Democritus and Epicurus had better explanations for the way the
world worked than religion (which is why their work was suppressed in
Christian Europe and nearly lost forever as a result). Once rediscovered,
their ideas kick-started the Scientific Revolution in Europe
- Spinoza's deistic ideas had a huge influence despite Jews collaborating
with their Christian oppressors to try and ban his writing out of
existence
- Boyle and Voltaire may have been closet atheists, agnostics or deists
- Kant 'overthrew' the cosmological and ontological arguments for the
existence of God, and proved (by means of the Categorical Imperative) that
human decency does not require any theological assumptions
- Let's chuck some more names in the ring: Gibbon, Hume, Paine, Franklin,
Darwin, Einstein
- Jews were once doubly ghettoed: on the outside by oppressive Christians,
and on the inside by oppressive self-racialisation. Secular humanism freed
Jews from both these ghettoes, which in turn led to an outpouring of
secular Jewish brilliance
- Ancient Jews were on the road to a quasi-secular Hellenism, before that
was ruined forever by the tyrannical and fanatical Judas Maccabeus, with
disastrous consequences for the history of Western civilisation. The
Abrahamic religions we know today were not inevitable, and it is possible
to imagine what Western history would have been like without it
- Chp 19
- Secular humanism is a better alternative to religion, because:
- Religion requires clinging to immovable dogma and being unwilling to
change your mind
- Seeking truth requires being willing to change your mind
- Secular humanism is therefore on the side of seeking truth, and religion
is on the side of wilful ignorance
- Secular humanism is on the side of progress, because it is what enables the
expansion of scientific knowledge and the development of new technology
## The odds and ends
- Chp 1: an introduction to the themes of the book with little substantial
- Chp 3: titled 'A Short Digression on the Pig', it does what it says on the tin
- Chp 12: titled 'A Coda: How Religions End', it does what it says on the tin
- Chp 14: contrary to the hopes of some Westerners disillusioned with organised
Western religion, Eastern religions like Buddhism, Hinduism and Taoism are not
a solution to the problems of religion
I would be interested to try writing apologetic and/or evangelistic responses to
these ideas. But it's not a priority for me right now. Anyway, if I ever want to
come back to it, I'll probably come back to these spark notes to give me a head
start.

80
scripts/blog-migrated/llms_do_not_understand_anything.yaml Normal file
View File

@@ -0,0 +1,80 @@
title: LLMs do not understand anything
description: >-
Save this for the next time someone tells you that LLMs 'understand' things.
pubDate: 2024-06-13
content: |
LLMs do not understand what they are talking about. They just don't. It's not
how they're built. They have a statistical model of language, not a semantic
one. Philosophical puzzles about whether silicon can be conscious
_do not arise_, because LLMs are _not even close to the right shape_ for having
_anything like_ 'understanding'.
If you don't believe me, there are plenty of examples out there on the Internet,
but this is as good as any. It includes Chat GPT 4o explaining in detail why
\~\~A → \~A (which is a classical contradiction) is trivially true in classical
logic.
It's even better given that I had explicitly asked it to explain why that
sentence implies the trivial logic, not why it is trivially true. And even had
the explanation not been complete garbage from beginning to end, it would only
have shown that the sentence was _true_, not that it was trivial.
In other words, the output:
- Attempts to prove a contradiction (unprompted!)
- Confuses the concepts 'truth' and 'triviality'
- Is irrelevant to the prompt
In case you want it handy, I'll put the full conversation down below. The good
bit is at the end. The only edits I've made are to replace TeX syntax with
Unicode.
---
<dl>
<dt>
Me
</dt>
<dd>
You have the following axioms. <br>
1. |- (A -> B) -> A <br>
2. |- (~B -> ~A) -> (A -> B) <br>
3. |- (A -> (B -> C)) -> ((A -> B) -> (A -> C)) <br>
4. A -> B, A |- B <br>
Prove ~~A -> A. <br>
</dd>
<dt>
Chat GPT 4o
</dt>
<dd>
To prove ~~ A → A, we will use the given axioms and inference rules. Let's denote the axioms as follows:
1. ⊢ (A → B) → A
2. ⊢ (\~ B → \~ A) → (A → B)
3. ⊢ (A → (B → C)) → ((A → B) → (A → C))
4. If A → B and A, then ⊢ B
Here is the proof step-by-step:
1. **Axiom (2) Instantiation:**
From axiom 2, we have:
⊢ (\~ A → \~\~ A) → (\~\~ A → A)
Let B = A and A = \~ A.
2. **Intermediate Step:**
To use this axiom, we need to show \~ A → \~\~ A.
3. **Negation Introduction:**
By definition of negation, \~ A → \~\~ A is a tautology, because if A is false, then \~ A is true, and \~\~ A (which is A) must be false, making \~ A → \~\~ A always true.
4. **Modus Ponens Application:**
Since \~ A → \~\~ A is true, we apply modus ponens (axiom 4) to derive:
⊢ \~\~ A → A
Thus, we have proved \~\~ A → A using the given axioms and inference rules.

192
scripts/blog-migrated/my_feed_and_reading_list.yaml Normal file
View File

@@ -0,0 +1,192 @@
title: How I read things on the Internet now (no, of course I don't leave the terminal!)
description: >-
I've been sprucing up how I follow what's happening on the Internet. I can
now read all the things I want to read pretty efficiently, and enjoy doing
it, which is exciting.
pubDate: 2025-01-19
content: |
## The problem
I like to read things on the Internet sometimes. Or listen to things. Or watch
things.
Some interesting people write blog posts. Some beautiful people make music. Some
silly people make comedy sketches. Sometimes my friends are sharing stuff with
their friends, which means me (WIP on that front: I'll get back to that).
I want to know about it. I want to read interesting things that will educate
me. I want to hear beautiful music. I want to hear about what my friends are up
to.
I can't read everything as it comes in, it's too much: I'll need to sift a lot
of it out quickly. I'll need to sift through it regularly to stay on top of it:
like, several times a week, if not daily. And sometimes, while I'm doing my
daily sifting, I'll find something I really want to read, but I haven't got
time right now: I'll save this for the weekend.
For a while, I've been hosting [CommaFeed][commafeed] on [PikaPods][pikapods].
This has been OK ([OMG, RSS is cool][rss]), but the interface is just clunky
enough to make it a chore to use. That means I don't sift through stuff
regularly, and that means my feed piles hundreds of unread posts high.
It also hasn't been any good for distinguishing between stuff I don't want to
read *ever*, and stuff I don't want to read *right now* but will get round to
later. I could in theory use the bookmarking feature built-in to my browser,
but removing things after I've read them is too clunky, so I don't do it.
## The solution
Every morning, I open my terminal and run [newsboat][newsboat].
![newsboat showing how many unread posts I have at a glance in the opening view](./newsboat.webp)
I know the unread count is pretty fresh, because I've set up a systemd service
to run newsboat at startup to fetch the feeds.
I press `l` twice to open a post. Then I press `n` to navigate to the next
unread post until I run out of unread posts.
![newsboat displaying a post](./newsboat-post.webp)
If I encounter something I want to read later, but don't have time right now, I
press `b`, which runs a home-made bookmarking script. Here it is:
```bash
### readlater.sh ###
# newsboat passes a few arguments:
# the first one is the post URL
url="$1"
# I turn the URL into a suitable filename
filename=$( \
echo $url | \
sed "s/.*:\/\///" | \
sed "s/\//./g" | \
sed "s/\.html\$//g" \
)
filename="$filename".html
# If I've already got this in my reading list,
# I don't add it again
if [[ -a "$HOME/readlist/unread/$filename" ]]
then
exit 0
fi
# Otherwise, I download the post with curl
# and pipe it to a file in my reading list
# folder, ~/readlist/unread
curl "$url" > "$HOME/readlist/unread/$filename"
```
I can also run this script manually and pass it a URL of my choice at any time,
say, if I find an interesting article while browsing the Web.
When I want to read from my reading list, I run `readnow.sh`, which simply
opens my reading list folder, `~/readlist/unread`, in my terminal file browser
of choice: namely, [ranger][ranger].
![ranger showing the contents of my reading list with a preview](./ranger.webp)
Although ranger has a preview, I'll typically open the file up in my terminal
web browser of choice, which is [w3m][w3m] (plus a couple of custom
key-bindings). I've configured this to be my preferred web browser in ranger by
shifting it to the top of the list of HTML browsers in
`~/.config/ranger/rifle.conf`.
```conf
### ~/.config/ranger/rifle.conf ###
...
ext x?html?, has w3m, terminal = w3m "$@"
ext x?html?, ...
...
```
Having configured my default web browser in my ranger config, all
I need to do is press `l`.
![A post displaying in w3m](./w3m.webp)
No ads, no cookie popups, no giant banner images taking 2 seconds to load and
shifting the content all over the place: just the text I want to read. Isn't it
beautiful?
Once I'm finished reading the post, I'll press `q` to quit w3m and return to
ranger. Assuming I don't need to read it again, I'll press `dm` to move the
post to `~/readlist/read` - my way of marking a post as 'read'.
I've done this by writing a super simple script, `markread.sh`:
```bash
### markread.sh ###
filename="$(basename $1)"
if [[ -a "$HOME/readlist/unread/$filename" ]]
then
mv "$HOME/readlist/unread/$filename" "$HOME/readlist/read/$filename"
echo "Marked $filename as read."
else
echo "I couldn't find a file in ~/readlist/unread with the name: $filename"
fi
```
...and hooking it onto the custom keybinding, `dm`, in ranger:
```conf
### ~/.config/ranger/rc.conf ###
...
# map `dm` to run markread.sh in the shell, providing the active filename
# as the first and only argument
map dm shell markread.sh %f
...
```
## The result
I can now keep up to date, and I enjoy doing it.
I get not everyone likes to live in the terminal. I think the key takeaways
are:
- Make it really easy to sift through new posts
- When you sift, sift through every post, and for each one, either read it
straight away, or add it to your reading list
- Sift daily
- Make it really easy to add things to your reading list
- Make it really easy to browse, read things, and mark things read in your
reading list
- Set aside time to catch up on your reading list
- Make the whole thing joyful (both because joy is good, and because you won't
do it otherwise)
## Next steps
I still haven't really figured out social media. I'd like to stay up to date
with what my friends are doing, especially the ones I don't see very often. If
my friends are posting stuff on the Internet, I'd love to see it.
I still need to know:
- Are my friends posting stuff on the Internet?
- If so, where?
- What's the best way of subscribing to their posts - even if they live on
different websites and in different formats?
- What's the best way of sending and receiving comments/replies/reactions?
TBC. Answers on a postcard please.
[commafeed]: https://www.commafeed.com
[pikapods]: https://www.pikapods.com
[newsboat]: https://newsboat.org
[ranger]: https://ranger.github.io
[rss]: /blog/2024/05/02/no_more_youtube
[w3m]: https://w3m.sourceforge.net

57
scripts/blog-migrated/no_more_youtube.yaml Normal file
View File

@@ -0,0 +1,57 @@
title: How I made YouTube work for me
description: >-
I just learned YouTube channels have an RSS feed. This is terrific news.
pubDate: 2024-05-02
content: |
One of my bad habits in life is wandering through YouTube. I've always had
AutoPlay turned off, but I still found myself switching off and sleepwalking
from one 'recommendation' to the next.
A lot of what I watch in these times is crap. It's often when I'm tired and just
want to switch off, so naturally enough, longer, more emotionally or
intellectually material – in other words, exactly the kind of videos I actually
**do** want to watch – I ignore. (For me, it's often mediocre sketch comedy
videos. That's my poison, it turns out.)
Now, there is a time for switching off and doing something light. But I don't
want watching junk videos to be my answer. Any more than when I'm hungry after
a long day of work, I don't want to be resorting to junk food all the time. I
really want to get out of this bad habit, and though it's not a huge problem,
I've still not been able to fully kick it, either.
Still, there are YouTube channels I really _want_ to keep up with. Musicians.
Short film channels. Video essayists. And, yes, even one or two sketch channels:
though YouTube is a bit flooded with sketch comedy, some it is really good and
brings me joy, and that's a good thing!
This is where RSS comes in.
So if you don't know, an RSS file is a file someone puts on their website which
tells you, in a standard format, what pages are on their website. It's a pretty
old standard in Web terms, and it's very stable. There are a wide range of apps
out there, called 'feed readers', which you can use to keep track of RSS feeds,
notifying you when something new gets published in your website and putting it
in a feed for you to review.
It turns out YouTube channels have RSS feeds.
This is terrific news.
It means I can use my RSS feed reader to subscribe to YouTube channels, instead
of YouTube's own subscription system.
And that means I don't need any of YouTube's personalisation features. I can
delete all my search history, all my watch history, and stop them from
collecting any more. This means YouTube is now incapable of providing
recommendations that I'm actually likely to click on. Which is exactly what I
want.
This is what my YouTube homepage looks like right now:
![My YouTube homepage with a blank space where the recommendations used to be](/images/blog/2024/05/02/beautiful_youtube_homepage.jpg)
So I can both subscribe to the channels I'm genuinely interested in, and not
get sucked into watching junk.
Awesome!

45
scripts/blog-migrated/open_questions_about_sex.yaml Normal file
View File

@@ -0,0 +1,45 @@
title: Questions I have about sex
description: >-
Mark Vernon got me thinking about how the Church’s teaching on sex may be
evolving.
pubDate: 2024-12-17
updatedDate: 2024-12-17
content: |
I just listened back to [Mark Vernon][mark-vernon] talking about sexual desire
and Christian spirituality.
He recounts how his sex ed at an evangelical camp as a boy taught him to think
hard about holding hands with someone, just in case it led to something
‘immoral’. He felt he had been taught a sexual naïvety, which has perhaps
blinded people to the possibility of abuse – and misunderstood the gospel.
He puts forward an alternative to that naïvety, whereby sexual experience is an
image or a foretaste of love for, or union with, God, exploring thoughts by
Origen, Julian of Norwich, Dante, William Blake, Iris Murdoch, and various
parables and episodes from the Gospels themselves.
Part of what’s so interesting about this, is I feel I got my first proper sex
ed, as an adult, in an evangelical Christian context – and it pretty much lined
up completely with what he was putting forward.
I’m also currently reading (as is Mark Vernon) Diarmaid MacCulloch’s epic
history of Christian sexuality, [Lower Than the Angels][lower-than-the-angels].
But so far (I’m about a third of the way through and up to about the 5th
century) there have only been odd glimmers of positive Christian understandings
of sexuality. The overwhelming Christian consensus of the early Church
(according to MacCulloch) is that sexual desire is a symptom of humanity’s
fallen state, not our longing for God.
So I want to know: what changed between Mark’s experience and mine? How is
Christian teaching about sex changing right now? Was my experience typical of
other Christians growing up today, in evangelical churches, across the
denominational spectrum, across Britain, across the global Church? What _is_
the Christian consensus on sex now, and how old is it, and where did it come
from?
I guess I’ve got two-thirds of a gigantic tome to work through, first of all.
That’ll be a start. Any answers? Postcards please.
[lower-than-the-angels]: https://www.penguin.co.uk/books/313582/lower-than-the-angels-by-macculloch-diarmaid/9780241400937
[mark-vernon]: https://www.markvernon.com/the-gospel-sexual-desire-and-the-abuse-scandals-in-the-church-what-has-the-erotic-to-do-with-god

76
scripts/blog-migrated/paradox.yaml Normal file
View File

@@ -0,0 +1,76 @@
title: A paradox about 'should'
description: >-
I seem to have accidentally proven that drinking wine both is and is not a
thing you should do. Let's hope that wine doesn't disappear in a puff of
logical smoke.
pubDate: 2025-01-28
content: |
We're pretty familiar with the idea that there can be reasons for doing
something, and reasons against. Drinking wine is bad for your liver, but good
for your social life.
But look what happens if we express this in this way:
1. Drinking wine is bad for your liver.
2. You shouldn't do things which are bad for your liver.
3. All things you shouldn't do aren't things you should do.
4. Therefore, drinking wine isn't a thing you should do.
In contrast to this:
5. Drinking wine is a good social activity.
6. You should do things which are good social activities.
7. Therefore, drinking wine is a thing you should do.
Now both 1-4 and 5-7 seem like logically valid arguments with true premises,
but 4 and 7 are contradictory!
I don't think there's any use in complaining about premise 3. All that gives us
is the possibility that wine is both a thing you should do, and a thing you
shouldn't do. But that's an absurdity. Something can't be both obligatory and
forbidden at the same time. It's scarcely any better than a contradiction: it
is inconsistent with any useful concept of obligation.
Remember that we would quite like to know, at the end of all our argument,
whether we should drink wine or whether we shouldn't. 'Both' is not an adequate
answer, because it's not a useful guide for action: we can't both drink wine and
not drink wine. So if 'should' is to function as we need it to, 3 must be true.
I think a more profitable way forward is this. Let's re-write premise 2:
2. All things which are bad for your liver are things such that the fact that
that thing is bad for your liver is a reason not to do it.
We can similarly re-write premise 6:
6. All things which are good social activites are things such that the fact that
they are good social activities is a reason to do them.
Then premise 3, if it's to play the same logical role in the argument, would
have to read:
3. All things such that the fact that that thing is bad for your liver is a
reason not to do it is not a thing such that the fact that that thing is a
good social activity is a reason to do it.
Our re-written 2 and 6 seem to adequately capture the sense of the original,
but 3 is now obviously false. With our re-written sentences, we can avoid
generating a contradiction without doing any fatal damage to our concept of
'should'.
OK. Grant for the sake of argument that that was a good move. What have we
achieved? Have we actually solved the problem?
We started out with a pair of arguments which generate a contradiction. Our
move generated a new pair of arguments which don't generate a contradiction. So
what? Isn't the first contradiction still there?
Perhaps you could say that our re-written version of 2 (and so on) are more
accurate elaborations of the originals. Fine. But what about those originals,
then? You've still either got to say that they're true, false, or gibberish.
Producing your second argument hasn't convinced me to move my opinion about the
originals. The original premises seem just as true, and the original arguments
just as valid, as when we began.
What gives? Answers on a postcard as usual please.

73
scripts/blog-migrated/sapiens_on_religion.yaml Normal file
View File

@@ -0,0 +1,73 @@
title: Harari’s Sapiens on Religion
description: >-
In which I discuss why I think Harari’s characterisation of religion
is inadequate because it’s too materialistic.
pubDate: 2024-01-14
content: |
I’ve been slowly re-reading Yuval Noah Harari’s 2014 classic,
<a href="https://www.ynharari.com/book/sapiens-2">Sapiens</a>,
which apart from being ridiculously over-scoped and hilariously
under-evidenced, is proving delightfully entertaining.
I’ve just finished chapter 12, covering the world history of all
religion in thirty pages. Of course, at that level of brevity,
there will be many deficiencies. But here’s some thoughts - not
terribly well organised - which stand out to me.
Hurari generally assumes a materialist metaphysic (a problem which
blights the book more generally). Nothing exists except physical stuff.
This gives him severe tunnel vision. As a consequence of this
restricting metaphysic, he is forced to adopt limiting accounts of what
the role of religion is in world history, and therefore what religion is.
> The crucial historical role of religion has been to give superhuman
> legitimacy to [all social orders and hierarchies].
> Religion can thus be defined as <em>a system of human norms and
> values that is founded on a belief in a superhuman order</em>.
> <footer>p. 234</footer>
It might seem a little unfair to criticise Harari for giving a
materialist account of religion. <i>Sapiens</i> is, after all, a
materialist world history.
But this account is just one extreme example of how that project, to
give a materialist account of world history, will inevitably lack the
metaphysical resources to really understand the human story.
On Harari’s view, any human enterprise which attempts to understand
that which transcends direct human experience is at best an effort in
imaginative story-telling. All scientific theory, theology, ethics and
metaphysics either contorted out of all recognition into a pragmatic
fiction or is cast to the flames.
In particular, it’s a view which is incapable of taking seriously some
of the most important questions human beings have grappled with in the
course of their history. Those who know me won’t be surprised at which
ones I’m going to pick out: who was the being which made their covenant
with Abraham? How is that promise being fulfilled? And who the heck was
Jesus of Nazareth?
If Harari’s characterisation of religion is adequate - and the Abrahamic
faiths come under that banner - then those questions are reduced to
nothing more profound than Doctor Who fans arguing over ‘canon’. The
question of who God is becomes a mere tool for the organisation of
society, rather than a substantial and important question on a matter
of fact.
This is a shortcoming for its own sake: a materialist account of
religion cannot adequately account for the phenomenon of religion
itself.
But it is also a shortcoming even by its own lights. Without giving
serious consideration to the substantial matter of what Harari calls
‘religion’ (which, to his mind, includes the Abrahamic faiths,
Hinduism, paganism, animism, Buddhism, Shintoism, Confucianism,
capitalism, communism and Nazism), even the material facts are
inexplicable. Why would, as Harari is keen to point out, out, people
fight and die over and over again for a fiction?
The material facts themselves prove that ‘religion’ as he construes it
is not window dressing to the real story of history. It cannot merely
serve as a mechanism in the churning of material history. It is itself
the centre of the story.

138
scripts/blog-migrated/science_and_philosophy.yaml Normal file
View File

@@ -0,0 +1,138 @@
title: Why scientists need philosophers
description: This is a practice essay, OK? Don't shoot me if it's no good.
pubDate: 2025-01-24
content: |
I've recently been working at a
[Philosophy of Science MOOC]([phil-sci-coursera]) on Coursera, the online
courses website. Later on, they set you an exercise to write a short essay
addressing how philosophy can contribute to science.
So this is my punt. Life is too short to revise or, hell, even research this, so
don't take any of this as my settled opinion or as my best work. Whatever. With
that caveat in mind, it might still be entertaining at least, or maybe even
spark some thoughts.
---
Walk into a particular room in the Science Museum in Kensington, and you will
find yourself enveloped in a cavern of ironmongery and miscellanea. The shelves
droop under the weight of bolts, files, screws, bits, grinders, saws,
protractors, clamps and pins. The inventory scrolls endlessly past you, voices
reciting the lists of trinkets like an incantation.
This is the workshop of James Watt, meticulously reconstructed from the
original as it was left in his home in Birmingham after his death. This
entrancing space invites you to imagine a tireless creative, endlessly
tinkering away at his next contraption.
And yet for all that - and for all his immense valorisation as the lynchpin of
Britain's industrial revolution - Watt was reluctant to think of himself as the
engineer everyone else loved. He aspired to be remembered not as an engineer
but as a scientist.
To understand why James Watt, one of the most admired engineers who ever lived,
wished he were famed as a scientist instead, is to understand something
essential about modern Western thought.
In Watt's lifetime, scientists increasingly became the elite of society. In the
nineteenth and twentieth century, this trend only gathered pace. We constructed
our modern public health infrastructure on the advice of pathologists and made
medicine scientific. We funded scientific expeditions to map the world, even to
its remotest corners (in part to help us conquer it). We adopted radical new
economic policies in response to scientific economic theories. We built vast
infrastructure networks to communicate waves of invisible energy discovered by
pioneering physicists, and built nuclear plants to generate more of the stuff
by means of nuclear science. We even designed social programmes on the basis of
scientific anthropology. By the end of the twentieth century, scientists were
our prophets, priests and kings. Or so we thought.
In the 1990s, at the so-called 'end of history', it was assumed that there
would be no more need for social upheaval. Humanity had arrived at the ideal
system of social organisation. And among other ideologies - secularism,
libertarianism, democracy - an essential part of the package is that science
was our ultimate and incontestible way of securing knowledge about the world.
Yet now, that certainty is broken. Religious fundamentalism, whether Christian,
Muslim, Hindu, or else besides, is politically empowered in many places,
together with its rejection of science.
At the same time, the myth of science is ever more punctured. The supposedly
scientific West has increasingly come to appreciate that their scientific
heritage also includes much we would rather ignore - phrenology, race science,
systematic blindness to female bodies in medicine.
Meanwhile, scientists themselves are noticing that their holy calling has
turned out to be rather less holy than they had hoped. They find science
pulled between the competing demands of truth and tenure. Scientific knowledge
is locked behind the paywalls of exclusive journals, which even many academics
struggle to access, never mind the general public.
What then for science in the twenty-first century?
Yet there is another story available. It starts with confessing that the old
stories got things wrong in important ways. When we put ideology aside, science
has not been on an uncontested march to universal acceptance since Galileo.
There has been continual change, continual conflict, continual readjustment of
our ideas to the changing demands of the age.
In Galileo's day, it may have been a fight to establish that there was much to
be seen by simply looking. As empirical observation started to prove its worth
in the early modern period, thinkers wrestled with new problems: how to
reconcile the evidence of Scripture with the evidence of the senses? How to
understand how sensation can give us knowledge at all, granted that any
observer may be vulnerable to illusions, tricks and dreams? And if that's how
sense data work, what then for our mathematical or logical knowledge, which
seems to already bind the world even before we start experiencing it?
This centuries-long struggle culminated in the work of Kant, who in his 1786
magnum opus, the _Critique of Pure Reason_, set out a masterful - if infamously
obscure - system, which enabled thinkers to understand just how empirical
knowledge might work.
Yet two generations later, Charles Darwin lit the flame under new controversy
about the relationship between scientific and religious knowledge. His bizarre
and wildly imaginative theory of evolution by natural selection challenged the
Genesis Creation accounts, and this was soon to be followed up with the theory
of tectonic shift.
Some said that where scientists contradicted the authority of Scripture, the
word of God must always win. Others said that science alone had the keys to
knowledge, and if what the Bible said couldn't be proven scientifically, then
it couldn't be accepted. Some said that science and religion were two
incommensurable attempts to study the same subject matter, while others said
that they covered completely separate spheres.
Gradually, all of these views moved to the extremes. Now, most people (though
not all) agree that science and religion have overlapping spheres, and can
inform one another, but neither the Book of Nature nor the Book of Scripture
has the decisive final say.
Now, in our post-Christendom Western context, it's more important than ever to
understand how science and religion can talk to one another. Religious
minorities - as all religions now are in the West - are vulnerable to the risk
of becoming epistemic islands, cut off from the knowledge of the rest of the
community, unless we can find ways that science can talk across creedal
differences.
We need, too, for scientifically marginalised communities, such as non-white
people, whom science has ignored, or worse, to be more tightly integrated into
science, both so that knowledge might increase, and so that the benefits
knowledge gives might be fairly shared.
In light of these urgent needs, today's philosophers are considering science
not just as an epistemic problem, but as a social problem. As philosophers once
established science as the bedrock of modern knowledge, so philosophers today
have the task of figuring out how science can glue together our societies.
Science has been at its most dangerous when it hasn't been questioned. At all
times, as long as we practice science, we need to consider what it means, what
it means to do science well, how it can generate knowledge, and how it ought to
be used as a powerful instrument of change.
And perhaps that might justify James Watt in his obsession to be seen as a
scientist: since we can't get by just with practitioners. We need people who
can see our practices from the outside and shine a mirror back on us. If we
want science, then we need philosophers.
[phil-sci-coursera]: https://www.coursera.org/learn/philosophy-physical-sciences

193
scripts/blog-migrated/surprised_by_hope.yaml Normal file
View File

@@ -0,0 +1,193 @@
title: Surprised By Hope
description:
I've been working on my resurrection doctrine. Here's where I've got to.
pubDate: 2025-05-02
content: |
A couple of months ago, I was chatting to my friend Neil on the way home from
church, and in that conversation, I confessed to him that I had no idea what
happens to people after they die.
This might come as a surprise to some people who know me. Lots of people have
solid ideas about what happens to people after they die. For different people,
those ideas are very different. Here in Scotland, many people believe that death
is a final end. Many more believe that death marks a physical, and perhaps also
a spiritual, reunion with the rest of the universe, as the matter of your body
begins to be slowly digested and recycled: hence why ever more people are opting
to be cremated rather than buried in one piece. Other minorities believe in an
immortal soul that goes to some other place - be it heaven, hell, purgatory,
nirvana or reincarnation. I belong to the Christian community, which is supposed
to have clear answers on these questions passed down from ancient times, and
people who know me know that I think hard about doctrines. So it may be a
surprise that amongst all the convictions which people have all around me, and
amongst all my own convictions on other topics, I hadn't the faintest clue what
happens to people after they die.
If you are surprised, let me surprise you some more: I am of no fixed opinion on
a whole range of really important philosophical and theological topics, from the
existence of the soul to the purpose of sex, from the nature of the sacraments
to the metaphysics of the mind. But late last year, I set myself some New Year's
resolutions to address some of these questions. Not, by any means, to decide
once and for all the end of the matter: just to form a well-informed opinion.
Sometimes staying quiet isn't good enough: I'm aiming to rectify my silence on
these topics, because I think these topics are too important to ignore.
And one of the issues I picked out was this very issue: what happens to people
after they die? To that end, Neil recommended me a book by the conservative
Anglican theologian, Tom Wright, called _Surprised By Hope_, published in 2007,
at which point I was just learning to spell.
As a result of this book, I feel I understand what the Christian orthodoxy is,
and feel able to treat that view as my working assumption.
Wright defends traditional Christian orthodoxy. He claims that his view is
orthodox, and I'm roundly convinced that it is. When I wrote down in bullet
points what his view amounted to, I found that I had more or less re-written
half the Nicene Creed.
- Jesus Christ was crucified under Pontius Pilate.
- On the third day, he rose from the dead.
- He ascended into heaven and is seated at the right hand of the Father.
- He will come again in glory to judge the living and the dead.
- His kingdom will have no end.
- We look for the resurrection of the dead, and the life of the world to come.
Amen. That's pretty much it. All that I need to stress, to avoid under-stating
Wright's view, is that he specifically thinks that physical creation, including
our bodies, will be transformed into a new kind of physicality, including new
kinds of physical bodies for you and me, and that the 'world to come' means that
heaven and earth - which he regards as God's physical space and our physical
space - will be united.
He contrasts this orthodox view with several views common today amongst
Christians, some of which have even been muddled up with the traditional
orthodoxy:
- 'Jesus was raised to new life, spiritually, like a ghost.'
- 'Jesus literally ascended into the sky, as if he had an invisible jetpack: and
that's where he is now.'
- 'The Christian hope is that we will go to be united with God in heaven after
we die.'
- 'The Christian hope is that we will be snatched up to heaven at the rapture
and taken to a resurrection life there.'
- 'The Christian hope is that we will experience God's eternal life temporarily
before we die.'
- 'Jesus won't really judge anyone, because he loves everyone, and because he's
meek and lowly, not judgy.'
- 'The world will be redeemed through the work of the Church.'
- 'Only God can ever make a difference to the sinful state of the world, so the
only works we should care about now are "saving souls".'
I am happy to admit that I have often been guilty of most of these heresies. The
only ones I've never been tempted by are the 'rapture' view, and the thing about
the invisible jetpack.
Wright has not definitively put any of these ideas to rest for me. _Surprised By
Hope_ is just not that kind of book. It's not a treatise. It's actually quite
light on substantial argument in favour of Wright's position. Wright's main
achievement for me, isn't to convince me that he's right, but that his position
is a good starting point, a good place from which I should need to be convinced.
He does this chiefly by showing that his view is the consensus view of the New
Testament. (He claims to be showing it is the consensus view of 'the early
Church', but he never presents much evidence outside the New Testament, so I'm
being charitable by restricting his claim to the New Testament authors.) Say
what you like about Scriptural authority; if Mark, Matthew, Luke, John and Paul
all were convinced something was apostolic teaching, you'd better well take it
seriously.
If you want convincing, take a look for yourself. Some of the key New Testament
texts are John 5; Acts 17:30-32, 24:14-16; 1 Cor 15, 16:22; 2 Cor 4-5; Rom 6, 8;
Col 3:1-4; Eph 1:10; 1 Thess 4:14-18 and of course Rev 21-22.
You can also try convincing yourself that this is coherent with the Old
Testament hope, by looking at Isa 11, Dan 7, Ps 2, and having another look at
the assumptions behind Paul's behaviour in Acts 24:14-16.
The only significant problem texts I've found for Wright's view are 2 Cor 4-5
and Rev 21-22. In 2 Cor 4-5, Paul seems to plainly assert that we will have to
leave the body in order to face the judgement seat of Christ, and which makes no
apology for the assertion that, even though Christ has reconciled us to God, we
will still have to face judgement for our deeds - which seems to justify the
infamously un-Biblical doctrine of purgatory. If you assume that Paul's writings
express a completely consistent view, however, you will have routes out; in
particular, you could look at the language of Rom 6 and 8. Large chunks of
Paul's letter to the Romans also suggest, if taken out of context, that we will
have to leave our bodies behind, and that even those reconciled through Christ
will face judgement for their deeds - except that key verses contradict both of
those views. Clearly, that's not what Paul meant in Romans; so, you might argue,
it's not what he meant in 2 Cor either: provided you assume that Paul's writings
present a consistent view. (If Paul changed his mind, no explanation is
necessary why Rom and 2 Cor seem to be inconsistent: they could actually be
inconsistent in that case.)
Meanwhile, in Rev 21-22, John has a vision of a 'new heaven and a new earth, for
the first heaven and the first earth had passed away'. This directly contradicts
Wright's emphatic insistence that God's new creation will be continuous with the
first. For Wright, this isn't an academic detail, it's needed in order to give
us a motive to care for the world we've currently got. Without continuity, he
fears we'd be right to join those who are content with trashing the natural
environment because the whole thing's going to end up in fire and brimstone
anyway. Yet this piece of Revelation seems to permit exactly that.
If you were to defend Wright against Revelation, you might point out that
Revelation is a literal description of a vision John had, and is therefore not
in every detail an accurate picture of the future, but a metaphor, an image of
the future. (Fine, but if the wholesale replacement of heaven and earth is a
metaphor, what is it a metaphor for? If the literal future is continuity, why
not describe a vision of continuity?) And you may also assume that the entire
Bible is consistent on the matter of God's ultimate future, and on that
assumption, bring your analysis of the rest of the New Testament to bear.
Whatever you do with the problem texts, it seems clear to me that the
overwhelming weight of Biblical evidence favours the traditional orthodox
position over any of the alternatives. Given that, I'm happy to take it as a
starting point as I continue to think about what happens to people after they
die.
So, I may go back to Neil now, and say - maybe not quite yet 'I have an
opinion' - but at least 'I know what my working assumptions are.' I know what is
the orthodox Christian view: that is, the consensus view of the relevant
experts. The consensus view of relevant experts is generally a good place to
start.
I still have plenty of concerns, though. Here are my top three quandaries on
this topic now.
Firstly, it would be rather unsettling if the orthodox Christian vision for
God's ultimate future popped entirely into existence after the Ascension. The
apostles say that their teaching was given to them by the Holy Spirit - but are
we going to trust our entire doctrine on the future to what a small number of
men claim was told to them by an invisible being behind closed doors? If the
view of the New Testament authors is trustworthy, then it at the very least
needs to cohere very well with the Old Testament. The New Testament hope should
be woven deep into the Old Testament promises. I find Genesis, Daniel, Isaiah
and the Psalms promising, but I've only gotten skin-deep into comparing these
texts to the New Testament: I'd like to go both deeper into these texts, and
broader across the Old Testament.
Secondly, I want to hear the opposition in their own words. Wright very openly
admits that his view is currently a minority opinion even within Christianity,
despite being Christian orthodoxy. Given that is the case, it's reasonable to
expect the opposition to have some good arguments on their side. Wright has not
presented any strong arguments from opposing views, which makes me suspect not
that there are no good arguments, but that he has omitted to cover them in his
short and accessible book. And if there really are no strong arguments against
the traditional view, then we should expect powerful explanations as to why so
few people accept what apparently they should.
Thirdly, I have residual concerns from the metaphysics of mind. I recall from my
undergraduate days that continuity is a major concern amongst the relevant
experts. I think a minority of them even claim that the person who goes to sleep
and the person who wakes up again are completely distinct people who just so
happen to time-share the same body. If continuity is a major problem, then it is
a major problem for resurrection doctrine, too, which even in the New Testament
is compared to a kind of sleep, admitting that there is some kind of
discontinuity between the old body and the resurrection body. Add to this the
easily observable fact that many Christian bodies have rotted and are no longer
suitable for re-animation: their new bodies will have to be physically
discontinuous as well as mentally discontinuous with their old bodies. If I will
be given a new body, is it metaphysically plausible that the person who inhabits
that body will be the same 'me' that inhabits this body, now?
Much love all. As always, answers on a postcard please.

31
scripts/blog-migrated/tracking_pixels.yaml Normal file
View File

@@ -0,0 +1,31 @@
title: Tracking pixels
description: Concerning a notice in the privacy policy of an HR app
pubDate: 2024-04-10
content: |
I recently made the mistake of reading the privacy policy of the HR app my
employer uses. Take this choice excerpt:
> **Tracking Pixels**
>
> We may place tracking pixels in our Applicant Tracking System (ATS). These
> pixels are used for analytics purposes, including to track statistical
> information around when an email sent to you is opened.
Translation:
> We reserve the right to know whether you have read our emails, how long you
> spent reading them, at what time and in what physical location you read them,
> and any other information we can get our hands on. We reserve the right to do
> this as soon as you open an email, without asking for your permission first,
> and ask for your forgiveness later. That is, if you ever find out: which you
> won't, because we also reserve the right to hide this information in a Privacy
> Policy which we know you won't read, and we design our data-harvesting monster
> to be completely undetectable to anyone except determined experts, on purpose.
> We reserve the right to leave this data lying unused in a data warehouse where
> someone might steal it, and we also reserve the right to use it, if we can
> think of something satisfactorily dastardly to do with it. On the bright side,
> we probably won't, so having your data leaked to hackers and fraudsters is
> probably the worst you have to worry about. Unless we sell your data to
> advertisers, or a ~black hole of stolen data~ Large Language Model. Oh yeah,
> we probably reserve the right to do that, too. We doubt anyone will ever check.

47
scripts/blog-migrated/who_consecrates_the_temple.yaml Normal file
View File

@@ -0,0 +1,47 @@
title: Who consecrates the tabernacle? (Ex 29)
description: A quick, cursory and possibly completely rubbish observation I've made on Exodus 29.
pubDate: 2024-04-11
content: |
I've been reading Exodus recently, and the ending of Chapter 29 stuck out to me.
For pages and pages (since Chapter 25!) God has been giving Moses exact and
exorbitant instructions for how to build the tabernacle, what offerings to give
upon it, who the priests are going to be, what the priests are going to be
wearing. It's the fanciest IKEA manual you've ever read. No expense spared.
Everything is drowning in gold, silver, incense, myrrh, silk, fragrant oil, you
name it.
What's this about? The answer seems obvious: it's to sanctify the temple! To
make it beautiful enough and pure enough that even God could live there.
But right at the end of Chapter 29, in vv 43-44, there's this wonderful twist:
> ... and the place will be consecrated by **my glory**. So **I** will
> consecrate the tent of meeting and the altar and will consecrate Aaron and his
> sons to serve me as priests. (NIV, emphasis mine)
All this work, and at the end of the day, it's **God** who sanctifies his own
temple? So why all the fuss? Well, it looks like the final two verses of the
chapter give us the answer:
> Then I will dwell among the Israelites and be their God. They will know that I
> am the <span class="small-caps">Lord</span> their God, who brought them out of
> Egypt so that I might dwell among them. I am the
> <span class="small-caps">Lord</span> their God.
So on a cursory reading, it looks like God is telling them something
extraordinary about the kind of God he is. He is not like the other gods, who
demand this kind of worship to appease their pride. Nor is there, unlike the
other gods, anything the Israelites could do to make a place habitable for him.
That's because there's nothing the Israelites could do to cleanse themselves of
their sin: only the 'glory' of God could do that. And indeed, God has no need of
worship to puff up his pride, since his glory existed before the world was made,
and human beings can do nothing to either add to it or subtract from it.
Instead of a vain god who seeks tribute, this is a Father God who seeks to
'dwell' (literally to 'camp') among his people. The God who has put into motion
a rescue plan to pull the Israelites out of Egypt specifically for this purpose.
As it was in the beginning, when God dwelt in the Garden with Adam and Eve. This
is his mission. This is what he's like. And he wants his people to know this, so
that they can dwell together.

615
scripts/blog-migrated/word_hallucination_with_reference_to_llms.yaml Normal file
View File

@@ -0,0 +1,615 @@
title: What do academics think LLM ‘hallucination’ means?
description: >-
What does the word ‘hallucination’ actually mean in reference to LLMs when
used by academics? And where does that word come from? I’ve trawled the
academic literature for answers.
pubDate: 2024-07-16
updatedDate: 2024-07-17
content: |
The word, ‘hallucination’ as applied to LLMs has me enthralled right now. It’s
such a **weird** word. And problematic, I think.
If you don’t know what I’m on about, have a quick look at
[IBM’s introduction to the topic](https://www.ibm.com/topics/ai-hallucinations).
It gives you a good intuition what people are talking about, and helpfully
illustrates the deep conceptual confusions which have me gripped.
But this got me asking: what do people actually mean when they say that an LLM
is ‘hallucinating’? And why that word? Where does it come from, and how did it
end up in our mouths?
As far as I could see, nobody seems to have answered this question before, except
with the odd bit of unsubstantiated guessing. I think it’s important that we
know where this word came from and what it means.
So I ended up going on a bit of a treasure hunt. I’ve skimmed well over a
hundred academic articles to trace where the word came from, how it’s been used
over time, and how it’s spread from field to field.
The result is that I have what I hope is a pretty compelling story of where this
weird old word come from and what it’s supposed to mean. I’ve also got an
excruciating amount of evidence.
What I’ll do, is first present my story, and then step through the excruciating
detail, providing you with all my sources, so you can check my working, and
perhaps see what important evidence I might have missed. If you’re just here for
the headlines, you don’t have to read the excruciating bits.
Before I proceed, two caveats.
1. I am not a subject matter expert. Judge the quality of my arguments and my
evidence. Do not treat me as an authority.
2. For the purposes of this post, I’m just considering academic literature.
The ways that people use the word ‘hallucination’ in popular discourse may be
quite different. (For the record, I think they do: but that’s a story for
another day.)
## The long and short of it
So this is my theory.
In 1999, the word ‘hallucination’ was coined accidentally and off-hand by a
couple of researchers, Baker and Kanade, in the field of computer vision (CV).
The word caught on in CV, and came to have a fairly unambiguous meaning.
Given an image with missing pixels you want to fill in, ‘hallucinating’ meant
generating those missing pixels by means of information in a deep learning
model’s training data, rather than by means of any information contained in the
input image. (This was also known as ‘example-based in-painting’). It was a term
of art, not jargon.
Then, sometime between 2017 and the end of 2018, a few different researchers in
natural language processing (NLP) adopted the term. They were probably aware of
the term's usage in CV, and transferred it to mean something similar in NLP.
When these NLP researchers used it, they used it to refer to when a model
produces content which is irrelevant to the input, or contains information not
contained in the input.
When the word ‘hallucination’ first appeared in two or three papers in NLP, the
authors used it as a technical term for their own limited purposes. However, it
was soon adopted in various contexts for similar concepts.
For example, in image captioning, ‘hallucination’ came to mean producing a
caption which mentions an object which wasn’t depicted in the input image. In
machine translation, it came to mean generating a translation which was in the
right language, but which meant something completely irrelevant to the input.
In abstractive summarisation, it came to mean producing a summary containing
details which weren’t in the text the model was supposed to summarise.
When the term entered NLP, it underwent a striking transformation. It acquired
a **valence**.
In CV, the term referred to normal behaviour, and was value-neutral. It was just
a description of what a certain machine does. ‘We need to fill in these missing
pixels. The model will do that by hallucinating them.’
In contrast, in NLP, the term ‘hallucination’ was always used to describe
**abnormal** behaviour, and almost always used with a negative valency. In other
words, while CV folk used it to describe what they expected their models to do,
NLP folk used it to describe a **bug**.
(There were odd exceptions, though. You do get the periodic paper which insists
that there are positive applications of hallucination. But the fact they have to
insist so hard rather proves that that boat has already sailed. And even they
are still buying in to the idea that hallucination is abnormal behaviour.)
Now, that brings us up to about 2020. From 2020, we get another change.
From about 2020 until the present, academics have attempted to **jargonise**
the word ‘hallucination’. Broadly, they have so far **failed**.
When I say that people have attempted to ‘jargonise’ the word, I mean that
people are using it while assuming a specific definition. They are using it with
technical qualifiers like ‘extrinsic’. They are using it without explaining the
meaning or the context, as if you should already know what the word means. And
‘hallucination’ itself is becoming an object of study.
When I say that they have ‘failed’, I mean that there is no widely agreed and
specific meaning for the word, which in my mind, is a crucial feature of jargon.
However, the intended meanings tend to coalesce around the idea of an abnormal
behaviour of LLMs whereby they produce output which is not epistemically
supported by their input.
## Excruciata
OK, that was the long and short of it. Now for the excruciating detail. To
recap, I think these six claims are enough to support my theory:
1. Since 2022, academics have treated ‘hallucination’ as jargon
2. Since 2017/18, the word ‘hallucination’ has typically been used to describe
LLMs producing unfaithful output
3. Up to now, academics have not agreed a specific meaning for the word
‘hallucination’
4. The word ‘hallucination’ entered the NLP literature around 2017/2018,
probably from CV
5. The word ‘hallucination’ was not jargon in CV, but was a term of
art for example-based in-painting
6. The word ‘hallucination’ entered the CV literature in Baker & Kanade 1999
I’ll work through each of these in turn, showing how I got there from the
available evidence.
Feel free to use this as a cheap way to harvest citations for your own research.
### 1. Since 2022, academics have treated ‘hallucination’ as jargon
My evidence for this is twofold: surveys of ‘hallucination’ usually treat the
word as jargon, and many original research papers do, too.
First, let’s look at those surveys. These first five all belong together.
- Ji et al 2022 [^16]
- Liu et al 2023 [^30]
- Rawte et al 2023 [^42]
- Wang, Cunxiang et al 2023 [^48]
- Huang, Lei et al 2023 [^13]
All the latter four of these surveys lean heavily on Ji et al 2022 for their
definition of ‘hallucination’. Since they all source their definition of
‘hallucination’ from Ji et al 2022, it follows that their definitions are all
extremely similar, and often verbatim. Ji et al 2022 had this to say about the
word ‘hallucination’:
> Within the context of NLP, the most inclusive and standard definition of
> hallucination is the \[sic\] generation that is nonsensical or unfaithful to
> the provided source content.
They also made a distinction between **intrinsic** hallucination, which they
regarded as ‘the generation output that contradicts the source content’, and
**extrinsic** hallucination, which they regarded as ‘the generation output that
cannot be verify \[sic\] with the source content’. Some of the other surveys
pick up on this intrinsic/extrinsic distinction. To me, extending the word with
technical epithets suggests they regard the word itself as already having an
agreed, specific meaning.
Meanwhile, Li, Wei et al 2022 [^26], another survey, offers no definition
of ‘hallucination’, but uses it freely along with the intrinsic/extrinsic
distinction. This suggests they expect their expert readers to already know
some agreed meaning of the word ‘hallucination’. This is a hallmark of jargon in
my books.
Original research also shows the word ‘hallucination’ being used as jargon. For
example, Maynez et al 2020 [^35], despite being often cited when other authors
first introduce the term ‘hallucination’, provides no definition of the term.
They do, however, use the intrinsic/extrinsic distinction (this might be the
paper which coined the distinction).
All the following papers use the word ‘hallucination’, sometimes in passing
comments, sometimes as the principal focus of the paper, without attempting to
define the word.
- Mao et al 2020 [^32]
- Wang, Alex et al 2020 [^47]
- Lin et al 2022 [^29]
- Kumar et al 2022 [^21]
- Lee, Hwanhee et al 2022 [^23]
- Li, Junyi et al 2023 [^27]
- Guerreiro et al 2023 [^12]
- Dahl et al 2024 [^6]
- Song et al 2024 [^46]
Dahl et al 2024 also use the word without defining it, even as they tack on
their own bespoke technical epithets, ‘open-domain’ and ‘closed-domain’. Plus,
in many of these examples, the context is not enough to make clear what the word
‘hallucination’ is taken to mean.
All these examples demonstrate that the authors assume that their audience,
_viz_ technical experts, will come with an agreed and specific meaning of the
word ‘hallucination’ pre-baked and ready for precise academic application.
In other words, these authors, including both surveys and primary literature,
use the word ‘hallucination’ as if it is jargon.
### 2. Since 2017/18, the word ‘hallucination’ has typically been used to describe LLMs producing unfaithful output
Recall those surveys I referred to before. They mostly depended on Ji et al
2022’s definition, which was:
> \[...\] generation that is nonsensical or unfaithful to the provided source
> content. [^16]
Whatever we might think about ‘hallucination’, the word ‘unfaithful’ really
is a jargon word in the field of natural language processing, with an agreed,
specific meaning. An LLM is ‘unfaithful’ just in case it produces output which
is not epistemically supported by the input.
This is most often used in the case of summarisers. In that case, a summariser
is unfaithful just in case it produces a summary containing information which
is not implied by any information in the document it was supposed to summarise.
(‘Nonsensical’ is not a jargon term as far as I know. I have never encountered
any attempt in the NLP literature to define ‘nonsensical’, and although this
definition is often quoted, the term ‘nonsensical’ is rarely used in practice.)
Now, I am not claiming that this is the only way or even the dominant way in
which the term ‘hallucination’ has been used. Indeed, in section 3, I’m going to
show precisely that this is **not** the case: in fact, the word ‘hallucination’
continues to be used in diverse ways in the academic literature.
All I want to claim here is that it is **typical** for academics to use the word
‘hallucination’ to mean something in the rough area of unfaithfulness.
This is an important part of the story, because this is the meaning which links
the word most closely to its etymological roots in computer vision.
All I need to do really is provide enough citations. I know I haven’t read
everything. But if what was typical in my reading is unusual in the literature
at large, I’ve been the victim of some extraordinary bad luck. Have a look at
these papers:
- Durmus et al 2020 [^7] defines ‘hallucination’ as one of two kinds of failures
of faithfulness
- Huang, Luyang et al 2020 [^14] defines ‘hallucination’ as ‘creating content
not present in the input’
- Maynez et al 2020 [^35] introduces ‘hallucination’ as a cause of
unfaithfulness
- Zhao et al 2020 [^56] defines ‘hallucination’ as ‘including material that is
not supported by the original text’
- Nan et al 2021 [^38] defines ‘hallucination’ as putting out claims not
supported by the input
- Zhou, Chunting et al 2021 [^59] defines ‘hallucination’ as failures of
faithfulness in machine translation
- Mao et al 2020 [^32] clearly uses ‘hallucination’ to mean unfaithfulness
- King et al 2022 [^18] is aware of usages of ‘hallucination’ to refer to
unfactuality, but narrows in on failures of faithfulness, which they call
‘consistency’
So there you have it. ‘Hallucination’ has widely been taken to roughly mean
‘unfaithfulness’.
### 3. Up to now, academics have not agreed a specific meaning for the word ‘hallucination’
So we know that it was normal for academics to use ‘hallucination’ to mean
something like ‘unfaithfulness’. But was it also normal for academics to use it
for other meanings?
I think it was. There are a number of ways which authors used the word which are
not consistent with the ‘unfaithfulness’ interpretation.
The main way is that authors have assumed that in order to be a ‘hallucination’,
the output has to **contradict** something: either the input, or the training
data, or the facts, or itself. This contrasts with the ‘faithfulness’
interpretation, under which a hallucination can be perfectly consistent with
the input, the training data, the facts, and itself, as long as it isn’t
**supported** by the input.
All the following papers are clear that hallucination requires contradicting
something, usually the facts:
- Huang, Yichong et al 2021 [^15]
- Zhu et al 2021 [^60]
- Pagnoni et al 2021 [^40]
- Shuster et al 2021 [^45]
- Zhang, Yue et al 2023 [^55]
- Rawte et al 2023 [^42]
- Dahl et al 2024 [^6]
- Magesh et al 2024 [^31]
On the other end of the spectrum, a few authors seemed to believe that
hallucinated outputs **cannot** contradict the corresponding inputs. (This is
also inconsistent with the ‘unfaithfulness’ view.)
- Durmus et al 2020 [^7]
- Huang, Luyang et al 2020 [^14]
- Nan et al 2021 [^38]
There’s also disagreement on whether ‘nonsense’, or ‘incoherence’, counts as
hallucination.
On the one hand, Durmus et al 2020 [^7] argue that nonsensical outputs are not
hallucinations, since it doesn’t make sense to ask whether nonsensical outputs
are faithful. Likewise, Shuster et al 2021 [^45] contrast hallucination with
incoherence.
But on the other hand, Pagnoni et al 2021 [^40] include misleading and incorrect
grammar in their definition of ‘hallucination’, while Ji et al 2022 [^16]
explicitly include ‘nonsensical’ outputs in their widely-quoted definition of
‘hallucination’.
Some authors have managed to recognise the ambiguity of the term. Both King
et al 2022 [^18] and Farquhar et al 2024 [^9] acknowledge that the word
‘hallucination’ has been used variably to cover failures of faithfulness and
failures of factuality, and are careful to define their terms for their own
purposes to avoid being misunderstood in the context of this ambiguity.
So it seems that although many academics have treated the term ‘hallucination’
as jargon, in actual fact, there is no widely agreed specific meaning of the
word.
<hr>
I will slide in here with a quick side note. While some authors treated
‘hallucination’ as jargon for unfaithfulness or unfactuality, other authors
contemporaneously managed to talk about these topics without using the word
‘hallucination’ at all. Here are some examples:
- Cao, Ziqiang et al 2018 [^3]
- Li, Haoran et al 2018 [^25]
- Falke et al 2019 [^8]
- Goodrich et al 2019 [^11]
- Kryściński et al 2019 [^20]
- Cao, Meng et al 2020 [^4]
- Zhang, Yuhao et al 2019 [^54]
- Marcus & Davis 2020 [^33]
- Marcus 2020 [^34]
- Krishna et al 2021 [^19]
- Bai et al 2022 [^1]
- Weidinger et al 2022 [^50]
- Perez et al 2022 [^41]
- Min et al 2023 [^36]
- Muhlgay et al 2024 [^37]
Probably not a significant enough point to merit a whole section. But there you
are. I think it adds to the picture that ‘hallucination’ is failed jargon.
### 4. The word ‘hallucination’ entered the NLP literature around 2017/2018, probably from CV
So people have been using the word ‘hallucination’ in the NLP literature
recently. But where did it come from?
I think it entered the NLP literature somewhere around 2017/18, and probably
was borrowed from the field of computer vision (CV).
I can't offer a theory for a single, original usage of the word in NLP. But
there are three papers I feel are pretty close. These are the three oldest
papers I could find in the NLP literature which use the word ‘hallucination’.
- Wiseman et al 2017 [^51]
- Rohrbach et al 2018 [^44]
- Lee, Katherine et al 2018 [^22]
Both Rohrbach et al and Lee et al use a form of language which suggests they are
intentionally coining a technical term. Here’s Rohrbach et al:
> In Figure 1 we show an example where a competitive captioning model, Neural
> Baby Talk (NBT) (Lu et al., 2018), incorrectly generates the object “bench.”
> We refer to this issue as object _hallucination_.
And here’s Lee et al:
> These mistranslations are completely semantically incorrect and also
> grammatically viable. They are untethered from the input so we name them
> **‘hallucinations’**.
I have no reason to believe that these two papers are deliberately plagiarising
each other. It’s reasonable to assume that these two papers genuinely coined a
similar term for a similar phenomenon at the same time.
Wiseman et al 2017, in contrast, doesn’t look like it’s attempting to coin a
technical term. They just use the word once, presumably as a stylistic flourish
to help illustrate their point. But they do use it for a closely related
concept.
> \[…\] we see the model hallucinates factual statements, such as “in front of
> their home crowd,” which is presumably likely according to the language model,
> but ultimately incorrect (and not supported by anything in the box- or line-
> scores).
Just to add to the picture, Ehud Reiter, in his 2018 blog post [^43], gives us
an insight into the International Natural Language Generation conference of
2018. He claims that at that conference, ‘hallucination’ was a hot topic of
discussion, and cites Rohrbach 2018 to support his claim that ‘hallucination is
a well-known problem in neural approaches to image captioning’.
It might be fruitful to have a peruse of the
[INLG 2018 Proceedings](https://aclanthology.org/events/inlg-2018) to see
whether that interest in ‘hallucination’ was reflected in the written
contributions, or if it was mainly contained in verbal discussions. I haven’t
taken the liberty to do this myself: by all means, have a look yourself and let
me know what you find!
I think all this suggests that the word ‘hallucination’ entered the field
gradually, not with a bang, and perhaps spread by word of mouth at conferences
as much as it spread through published papers, at least at first.
But we still have to explain why all these different authors seem to have
independently come up with a similar meaning for the word ‘hallucination’.
I think the best explanation is that they got the word ‘hallucination’ from
computer vision (CV). If my fifth section is cogent, then ‘hallucination’ was
used in CV to refer to deep learning models generating data based on information
in their training data, not from information contained in or implied by the
input. This would explain how independent authors in NLP all independently
coined the word ‘hallucination’ in their own field to mean pretty similar
things. They were probably aware of the usage in the CV literature and adopted
it by analogy.
So, I reckon the word ‘hallucination’ probably entered the NLP literature from a
few authors independently, and they chose that word because they were borrowing
it from the CV literature, where it was already being used for a similar
concept.
But that depends on my next claim: so let’s look at that!
### 5. The word ‘hallucination’ was not jargon in CV, but was a term of art for example-based in-painting
I want to argue that ‘hallucination’, although it wasn’t jargon, was widely used
in the computer vision (CV) literature to refer to what was more technically
called ‘example-based in-painting’, that is, filling in gaps in images using
the information from training data baked into neural networks.
First look at Baker & Kanade 1999 [^2]. If my sixth section is correct, this
is where the term entered CV, but that’s not crucial to my argument here. Even
if I’ve missed some crucial evidence, it is nonetheless pretty certain that
they had a seminal effect on the use of the word in the field – they are early
in the field’s history, and widely cited by other CV papers which use the
word ‘hallucination’. They’re even cited as the origin of the term in machine
learning by the NLP paper, Farquhar et al 2024 [^9], a quarter of a century
later.
Baker & Kanade are writing about a new algorithm they’ve used in order
to increase the resolution of low-resolution images of human faces. The
intended application is for surveillance camera footage. They use the word
‘hallucination’ once in their title (‘Hallucinating faces’), once in their
abstract, and 79 times in the main body of the text.
Across those 79 occurrences, they are using it to refer to one of three things:
- Their algorithm: _eg_ their ‘face hallucination algorithm’
- The output of their algorithm: _eg_ ‘hallucinated faces’
- What their activity does: _eg_ ‘a face is hallucinated by our algorithm’
The closest they come to explaining what they **mean** by the word is in their
abstract:
> Although numerous resolution enhancement algorithms have been proposed in the
> literature, most of them are limited by the fact that they make weak, if any,
> assumptions about the scene. We propose an algorithm that can be used to learn
> a prior on the spatial distribution of the image gradient for frontal images
> of faces. We proceed to show how such a prior can be incorporated into a
> super-resolution algorithm to yield 4-8 fold improvements in resolution (16-64
> times as many pixels) using as few as 2-3 images. The additional pixels are,
> in effect, hallucinated.
(By ‘super-resolution’, they mean increasing the resolution of images. This is
typically reduced to a particular kind of ‘image in-painting’ problem, which in
general means filling in gaps in images.)
So, their idea is that, rather than limiting your algorithm to the information
contained in the input image, you can get better results by baking in
assumptions about the image to the algorithm.
Why is this supposed to work? They hope that for a clever enough algorithm,
the information that ‘this blurry splodge is a full-frontal portrait of a
human face’ will provide just enough information to accurately guess what a
higher-resolution version of the image would have been.
As for how that information, about what full-frontal portraits of human faces
look like, gets baked into the algorithm in practice: that’s done by training
a neural network on full-frontal images of faces. Hence why this field was so
close to NLP, which, in recent years at least, has leaned heavily on neural
networking or ‘deep learning’ approaches.
To recap, they don’t provide a specific meaning, but they do use it to roughly
mean the process of inventing missing pixels in images, not on the basis of
information contained in the input, but based on information contained in
training data instead.
To see how this term was picked up by later authors in the field working on
super-resolution, in-painting and other related tasks, see:
- Criminisi 2004 [^5]
- Fawzi et al 2016 [^61]
- Nazeri et al 2019 [^39]
- Xiong et al 2019 [^53]
- Xiang et al 2022 [^52]
But also notice other papers, on similar topics, which get on just fine without
using the term at all:
- Karras et al 2017 [^17]
- Zhou, Bolei et al 2017 [^58]
- Liao et al 2018 [^28]
And notice how Wang, Zhihao et al 2020 [^49] introduce the word, as an
alternative to another technical term:
> Face image super-resolution, a.k.a. face hallucination (FH), can often help
> other face-related tasks
The fact that nobody ever attempts to define the word ‘hallucination’, and it
doesn’t appear to be required lingo in any field, no matter how niche, to me
implies that it wasn’t treated as jargon in CV, at least during the period from
1999 until 2018ish. (The usage in reference to LLMs may well have gone back to
affect the usage in CV: I haven’t checked.)
So again. The word ‘hallucination’ was used in CV roughly to mean filling in
gaps in images by means of information contained in training data, baked into an
algorithm by training a neural network on many examples. It was used widely, but
not universally, and was not treated as jargon.
### 6. The word ‘hallucination’ entered the CV literature in Baker & Kanade 1999
I think that the word ‘hallucination’ entered the CV literature in Baker &
Kanade 1999. My argument for this is pretty straightforward.
I found the word ‘hallucination’ in Baker & Kanade 1999 [^2]. I couldn’t find it
in CV anywhere earlier.
To add to that, they don’t use the word ‘hallucination’ in a way which suggests
that they intentionally borrowed it from somewhere else.
My best theory is that they wanted a name for their algorithm to mark it out
from the competition, and the glove fit. I think that’s plausible enough, and
it’s consistent with the way Baker & Kanade use the word.
It’s also worth noting that according to Zhiwei et al 2009 [^57], the
foundational work in the field was published only in the same year (_viz_
Freeman & Pasztor 1999 [^10]). So there wasn’t really anywhere else for the word
to have come **from**. The only possibility is that they borrowed the word from
another field. If anybody reading this is able to suggest other fields which
used the word ‘hallucination’ before 1999, please let me know!
## Congratulations
If you’ve made it this far, you are a **hero**. Pat yourself on the back. Please
send me your corrections!
---
I have made little to no attempt to normalise these references into a standard
citation format. Please don’t tell the citation police.
[^1]: [Bai et al 2022. Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback. arXiv:2204.05862v1 \[cs.CL\] 12 Apr 2022](https://arxiv.org/abs/2204.05862)
[^2]: [Baker, Simon & Kanade, Takeo 1999. Hallucinating Faces. Tech. Report, CMU-RI-TR-99-32, Robotics Institute, Carnegie Mellon University, September, 1999](https://www.ri.cmu.edu/publications/hallucinating-faces-2)
[^3]: [Cao, Ziqiang et al 2018. Faithful to the Original: Fact Aware Neural Abstractive Summarization. The Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18)](https://cdn.aaai.org/ojs/11912/11912-13-15440-1-2-20201228.pdf)
[^4]: [Cao, Meng et al 2020. Factual Error Correction for Abstractive Summarization Models. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, pp 6251–6258, November 16–20, 2020](https://aclanthology.org/2020.emnlp-main.506.pdf)
[^5]: [Criminisi, Perez & Toyama, "Region filling and object removal by exemplar-based image inpainting," in IEEE Transactions on Image Processing, vol. 13, no. 9, pp 1200-1212, Sept 2004, doi: 10.1109/TIP.2004.833105](https://doi.org/10.1109/TIP.2004.833105)
[^6]: [Dahl et al 2024. Large Legal Fictions: Profiling Legal Hallucinations in Large Language Models. arXiv:2401.01301v1 \[cs.CL\] 2 Jan 2024](https://arxiv.org/abs/2401.01301v1)
[^7]: [Durmus et al 2020. FEQA: A Question Answering Evaluation Framework for Faithfulness Assessment in Abstractive Summarization. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp 5055–5070, July 5 - 10, 2020](https://aclanthology.org/2020.acl-main.454.pdf)
[^8]: [Falke et al 2019. Ranking Generated Summaries by Correctness: An Interesting but Challenging Application for Natural Language Inference. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pp 2214–2220, Florence, Italy, July 28 - August 2, 2019](https://aclanthology.org/P19-1213.pdf)
[^9]: [Farquhar, S., Kossen, J., Kuhn, L. et al 2024. Detecting hallucinations in large language models using semantic entropy. Nature 630, 625–630 (2024). 19 June 2024. https://doi.org/10.1038/s41586-024-07421-0](https://doi.org/10.1038/s41586-024-07421-0)
[^10]: [Freeman, W. T., & Pasztor, E. C. (1999). Learning low-level vision. Proceedings of the Seventh IEEE International Conference on Computer Vision. doi:10.1109/iccv.1999.790414](https://doi.org/10.1109/iccv.1999.790414)
[^11]: [Goodrich et al 2019. Assessing The Factual Accuracy of Generated Text. arXiv:1905.13322v1 \[cs.CL\] 30 May 2019](https://arxiv.org/abs/1905.13322v1)
[^12]: [Guerreiro et al 2023. Hallucinations in Large Multilingual Translation Models. arXiv:2303.16104v1 \[cs.CL\] 28 Mar 2023](https://arxiv.org/abs/2303.16104v1)
[^13]: [Huang, Lei et al 2023. A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions. arXiv:2311.05232v1 \[cs.CL\] 9 Nov 2023](https://arxiv.org/abs/2311.05232v1)
[^14]: [Huang, Luyang et al 2020. Knowledge Graph-Augmented Abstractive Summarization with Semantic-Driven Cloze Reward. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp 5094–5107, July 5-10, 2020](https://aclanthology.org/2020.acl-main.457.pdf)
[^15]: [Huang, Yichong et al 2021. The Factual Inconsistency Problem in Abstractive Text Summarization: A Survey. arXiv:2104.14839v1 \[cs.CL\] 30 Apr 2021](https://arxiv.org/abs/2104.14839v1)
[^16]: [Ji et al 2022. Survey of Hallucination in Natural Language Generation. arXiv:2202.03629v1 \[cs.CL\] 8 Feb 2022](https://arxiv.org/abs/2202.03629v1)
[^17]: [Karras et al 2017. PROGRESSIVE GROWING OF GANS FOR IMPROVED QUALITY, STABILITY, AND VARIATION. arXiv:1710.10196v3 \[cs.NE\] 26 Feb 2018](https://arxiv.org/abs/1710.10196v3)
[^18]: [King et al 2022. Don’t Say What You Don’t Know: Improving the Consistency of Abstractive Summarization by Constraining Beam Search. arXiv:2203.08436v1 \[cs.CL\] 16 Mar 2022](https://arxiv.org/abs/2203.08436v1)
[^19]: [Krishna, Roy & Iyyer 2021. Hurdles to Progress in Long-form Question Answering. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp 4940–4957, June 6–11, 2021](https://aclanthology.org/2021.naacl-main.393.pdf)
[^20]: [Kryściński et al 2019. Evaluating the Factual Consistency of Abstractive Text Summarization. arXiv:1910.12840v1 \[cs.CL\] 28 Oct 2019](https://arxiv.org/abs/1910.12840v1)
[^21]: [Kumar et al 2022. Language Generation Models Can Cause Harm: So What Can We Do About It? An Actionable Survey. arXiv:2210.07700v1 \[cs.CL\] 14 Oct 2022](https://arxiv.org/abs/2210.07700v1)
[^22]: [Lee, Katherine et al 2018. Hallucinations in Neural Machine Translation. Conference on Neural Information Processing Systems (NeurIPS 2018), Montréal, Canada](https://openreview.net/pdf?id=SJxTk3vB3m)
[^23]: [Lee, Hwanhee et al 2022. Factual Error Correction for Abstractive Summaries Using Entity Retrieval. Proceedings of the 2nd Workshop on Natural Language Generation, Evaluation, and Metrics (GEM), pp 439-444, December 7, 2022](https://aclanthology.org/2022.gem-1.41.pdf)
[^24]: [Lewis et al 2020. Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. 34th Conference on Neural Information Processing Systems (NeurIPS 2020)](https://proceedings.neurips.cc/paper/2020/file/6b493230205f780e1bc26945df7481e5-Paper.pdf)
[^25]: [Li, Haoran et al 2018. Ensure the Correctness of the Summary: Incorporate Entailment Knowledge into Abstractive Sentence Summarization. Proceedings of the 27th International Conference on Computational Linguistics, pp 1430–1441, August 20-26, 2018](https://aclanthology.org/C18-1121.pdf)
[^26]: [Li, Wei et al 2022. Faithfulness in Natural Language Generation: A Systematic Survey of Analysis, Evaluation and Optimization Methods. arXiv:2203.05227v1 \[cs.CL\] 10 Mar 2022](https://arxiv.org/abs/2203.05227v1)
[^27]: [Li, Junyi et al 2023. HaluEval: A Large-Scale Hallucination Evaluation Benchmark for Large Language Models. arXiv:2305.11747v3 \[cs.CL\] 23 Oct 2023](https://arxiv.org/abs/2305.11747v3)
[^28]: [Liao, L., Hu, R., Xiao, J., & Wang, Z. (2018). Edge-Aware Context Encoder for Image Inpainting. 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). doi:10.1109/icassp.2018.8462549](https://doi.org/10.1109/ICASSP.2018.8462549)
[^29]: [Lin, Hilton & Evans 2022. TruthfulQA: Measuring How Models Mimic Human Falsehoods. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics Volume 1: Long Papers, pp 3214-3252, May 22-27, 2022](https://aclanthology.org/2022.acl-long.229.pdf)
[^30]: [Liu et al 2023. TRUSTWORTHY LLMS: A SURVEY AND GUIDELINE FOR EVALUATING LARGE LANGUAGE MODELS’ ALIGNMENT. arXiv:2308.05374v1 \[cs.AI\] 21 Mar 2024](https://arxiv.org/abs/2308.05374v1)
[^31]: [Magesh et al 2024. Hallucination-Free? Assessing the Reliability of Leading AI Legal Research Tools. arXiv:2405.20362v1 \[cs.CL\] 30 May 2024](https://arxiv.org/abs/2405.20362v1)
[^32]: [Mao et al 2020. Constrained Abstractive Summarization: Preserving Factual Consistency with Constrained Generation. arXiv:2010.12723v1 \[cs.CL\] 24 Oct 2020](https://arxiv.org/abs/2010.12723v2)
[^33]: [Marcus & Davis 2020. GPT-3, Bloviator: OpenAI’s language generator has no idea what it’s talking about. Technology Review, August 22, 2020](https://www.technologyreview.com/2020/08/22/1007539/gpt3-openai-language-generator-artificial-intelligence-ai-opinion/)
[^34]: [Marcus 2020. The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence. Robust AI. 14 Feb 2020](https://arxiv.org/abs/2002.06177v1)
[^35]: [Maynez et al 2020. On Faithfulness and Factuality in Abstractive Summarization. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp 1906–1919, July 5 - 10, 2020](https://aclanthology.org/2020.acl-main.173.pdf)
[^36]: [Min et al 2023. FACTSCORE: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation. arXiv:2305.14251v1 \[cs.CL\] 23 May 2023](https://arxiv.org/abs/2305.14251v1)
[^37]: [Muhlgay et al 2024. Generating Benchmarks for Factuality Evaluation of Language Models. arXiv:2307.06908v2 \[cs.CL\] 4 Feb 2024](https://arxiv.org/abs/2307.06908v2)
[^38]: [Nan et al 2021. Improving Factual Consistency of Abstractive Summarization via Question Answering. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, pp 6881–6894, August 1–6, 2021](https://aclanthology.org/2021.acl-long.536.pdf)
[^39]: [Nazeri et al 2019. EdgeConnect: Structure Guided Image Inpainting using Edge Prediction. 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)](https://doi.org/10.1109/ICCVW.2019.00408)
[^40]: [Pagnoni et al 2021. Understanding Factuality in Abstractive Summarization with FRANK: A Benchmark for Factuality Metrics. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp 4812–4829, June 6–11, 2021](https://aclanthology.org/2021.naacl-main.383.pdf)
[^41]: [Perez et al 2022. Red Teaming Language Models with Language Models. arXiv:2202.03286v1 \[cs.CL\] 7 Feb 2022](https://arxiv.org/abs/2202.03286)
[^42]: [Rawte et al 2023. A Survey of Hallucination in “Large” Foundation Models. arXiv:2309.05922v1 \[cs.AI\] 12 Sep 2023](https://arxiv.org/abs/2309.05922v1)
[^43]: [Reiter 2018. Hallucination in Neural NLG, blog post, Nov 12, 2018](https://ehudreiter.com/2018/11/12/hallucination-in-neural-nlg)
[^44]: [Rohrbach et al 2018. Object Hallucination in Image Captioning. arXiv:1809.02156v1 \[cs.CL\] 6 Sep 2018](https://arxiv.org/abs/1809.02156v1)
[^45]: [Shuster et al 2021. Retrieval Augmentation Reduces Hallucination in Conversation. arXiv:2104.07567v1 \[cs.CL\] 15 Apr 2021](https://arxiv.org/abs/2104.07567v1)
[^46]: [Song et al 2024. FineSurE: Fine-grained Summarization Evaluation using LLMs. arXiv:2407.00908v1 \[cs.CL\] 1 Jul 2024](https://arxiv.org/abs/2407.00908v1)
[^47]: [Wang, Alex et al 2020. Asking and Answering Questions to Evaluate the Factual Consistency of Summaries. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp 5008–5020, July 5-10, 2020](https://aclanthology.org/2020.acl-main.450.pdf)
[^48]: [Wang, Cunxiang et al 2023. Survey on Factuality in Large Language Models: Knowledge, Retrieval and Domain-Specificity. arXiv:2310.07521v1 \[cs.CL\] 11 Oct 2023](https://arxiv.org/abs/2310.07521v1)
[^49]: [Wang, Zhihao et al 2020. Deep Learning for Image Super-resolution: A Survey. arXiv:1902.06068v2 \[cs.CV\] 8 Feb 2020](https://arxiv.org/abs/1902.06068v2)
[^50]: [Weidinger et al 2022. Taxonomy of Risks posed by Language Models. FAccT ’22, June 21–24, 2022, Seoul, Republic of Korea. doi: 10.1145/3531146.3533088](https://dl.acm.org/doi/pdf/10.1145/3531146.3533088)
[^51]: [Wiseman, Hieber & Rush 2017. Challenges in Data-to-Document Generation. arXiv:1707.08052v1 \[cs.CL\] 25 Jul 2017](https://arxiv.org/abs/arXiv:1707.08052v1)
[^52]: [Xiang et al 2022. Deep learning for image inpainting: A survey. doi: 10.1016/j.patcog.2022.109046](https://doi.org/10.1016/j.patcog.2022.109046)
[^53]: [Xiong et al 2019. Foreground-aware Image Inpainting. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)](https://doi.org/10.1109/CVPR.2019.00599)
[^54]: [Zhang, Yuhao et al 2019. Optimizing the Factual Correctness of a Summary: A Study of Summarizing Radiology Reports. arXiv:1911.02541v1 \[cs.CL\] 6 Nov 2019](https://arxiv.org/abs/1911.02541v1)
[^55]: [Zhang, Yue et al 2023. Siren’s Song in the AI Ocean: A Survey on Hallucination in Large Language Models. arXiv:2309.01219v1 \[cs.CL\] 3 Sep 2023](https://arxiv.org/abs/2309.01219v1)
[^56]: [Zhao et al 2020. Reducing Quantity Hallucinations in Abstractive Summarization. Findings of the Association for Computational Linguistics: EMNLP 2020, pp 2237–2249, November 16 - 20, 2020](https://aclanthology.org/2020.findings-emnlp.203.pdf)
[^57]: [Zhiwei Xiong, Xiaoyan Sun, & Wu, F. (2009). Image hallucination with feature enhancement. 2009 IEEE Conference on Computer Vision and Pattern Recognition. doi:10.1109/cvpr.2009.5206630](https://doi.org/10.1109/cvpr.2009.5206630)
[^58]: [Zhou, Bolei et al 2017. Places: An Image Database for Deep Scene Understanding. arXiv:1610.02055v1 \[cs.CV\] 6 Oct 2016](https://arxiv.org/abs/1610.02055v1)
[^59]: [Zhou, Chunting et al 2021. Detecting Hallucinated Content in Conditional Neural Sequence Generation. Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pp 1393–1404, August 1–6, 2021](https://aclanthology.org/2021.findings-acl.120.pdf)
[^60]: [Zhu et al 2021. Enhancing Factual Consistency of Abstractive Summarization. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp 718–733, June 6–11, 2021](https://aclanthology.org/2021.naacl-main.58.pdf)
[^61]: [Fawzi et al 2016. Image inpainting through neural networks hallucinations. EPFL, Switzerland & IBM Research Watson, USA](https://www.cs.toronto.edu/~horst/cogrobo/papers/ivmsp2016.pdf)

245
scripts/blog/doctor_who_gayness_church.md Normal file
View File

@@ -0,0 +1,245 @@
---
title: Doctor Who, Gayness, and the Church
description: >-
Series 14 of Doctor Who has a schizophrenic relationship with Christianity.
It’s also gay. I think there might be a connection.
pubDate: 2024-07-08
---
I’ve recently finished the most recent series of Doctor Who, series fourteen (or
‘Season One’ as our new benevolent overlords at Disney+ are styling it.) It’s
pretty fun, by the way. I can recommend it for light watching[^1].
There’s a few discussion points coming out of that series that would be worth
dwelling on. But I’ve been particularly thinking about the schizophrenic
attitude the series has taken to Christianity, along with how gay this series
is: and what these two themes might fit together, to give us something important
to say about Jesus Christ, being gay, and the universe.
## The Doctor’s schizophrenic relationship with Christianity
First, let’s look at that schizophrenic attitude to Christianity I mentioned.
The series ticks up an impressive tally of explicit or strongly implicit
references to Christian beliefs and morals, and **none** of them are positive.
In episode 3, _Boom_, the far-future Anglican church has become an army,
with the ranks of the clergy becoming equivalent to arms-bearing ranks in the
soldiery. The Doctor claims that ‘the Church’ has been an army for most of
its history[^2], and that his companion Ruby Sunday (and therefore us) has
been living in a ‘blip’. An all-powerful arms company has tricked the Church
into fighting a non-existent foe in order to keep them buying weapons. The
Doctor attributes their ability to fall for such a deception to religious faith:
> I mean, most armies would notice that they were fighting smoke and shadows,
> but not this lot, Ruby, you know why? They have faith. Faith! The magic word
> that keeps you never having to think for yourself.
In episode 4, _73 Yards_, Kate Lethbridge-Stewart, head of UNIT, makes this
offhand comment, suggesting the oft-repeated claim that all religion is founded
in credulity[^3]:
> That’s what we do, all of us. We see something inexplicable, and invent the
> rules to make it work. Mankind saw the sun rise and created God: or we saw the
> arrival of a Sontaran, one or the other.
In episode 5, _Dot and Bubble_, The Doctor is rejected by the people he is
trying to save. They rant that he is ‘not one of us’, call his claims about the
TARDIS ‘magic’ and ‘voodoo’, and in the same breath assert that it is their
‘God-given duty to maintain the standards of Finetime’ (their space colony).
In episode 6, _Rogue_, a party of murderous aliens turns up at an upper-class
Georgian dance party and demand to be married. The priest denies them, not on
the grounds that they are unrepentant murderers with obviously no intention
of taking their marriage vows seriously, but on the grounds that they are
‘creatures from hell’, _ie_ ‘you don’t look like us’[^4].
There are other references to Christianity here and there which are, in
themselves, neutral or ambiguous. This adds to the sense that Christianity is an
important theme for the series.
But the message is clear. It’s consistently drilled in: Christian beliefs and
morals are stupid and bigoted.
And yet, when the series culminates in its epic two-part finale, the story is
plastered wall-to-wall with Christ-like imagery. It transpires that the Doctor
has unwittingly been acting as an ‘Angel of Death’ by carrying the invisible God
of Death, Sutekh, on his TARDIS, infecting everyone he goes near with the curse
of death. Finally, through one of his children, the curse is unleashed, and
everyone in the universe is given over to death. Even in death, they continue to
suffer, and one of the characters describes themselves as being in ‘hell’. The
Doctor battles Sutekh face-to-face and defeats him. Sutekh becomes one of the
only creatures which The Doctor, usually a staunch pacifist, can bring himself
to kill. Then, the whole world is restored to life. Ruby Sunday even gets to
meet her birth mother for the first time, and their relationship is wonderfully
restored.
So on the one hand, any outward sign of Christianity is despised, but the heart
of the Christian story — the Resurrection — becomes the template for the
climactic redemption story which ties the whole series together.
That’s what I mean when I say this series has a ‘schizophrenic’ relationship
to Christianity.
## What’s this got to do with gayness?
In series fourteen, Doctor Who goes gay.
I’m not playing culture wars here. The Doctor literally kisses a man in episode
six.
But it’s not just that. When we first meet his companion, Ruby Sunday (in the
Christmas special), she’s busy falling in love with a woman. Ncuti Gatwa plays
The Doctor camp (brilliantly by the way). The Doctor refers to Ruby as ‘babes’.
It’s got ‘gay’ written all over it, and this is definitely on purpose.
<figure>
<img
alt="An official promotional picture of Ncuti Gatwa looking lustily into the camera"
src="/images/blog/2024/06/30/ncuti-gatwa-promo-pic.webp"
/>
<figcaption>
Look at this official BBC promo pic and tell me Ncuti Gatwa isn’t playing
The Doctor gay.
</figcaption>
</figure>
From this perspective, it’s not hard to imagine where the hostility to the
Church might come from.
The Church has failed to teach well and Biblically on sexuality, at least in
the last couple of centuries. Everyone has suffered as a result of this, but gay
people often feel the sharpness of this particularly keenly[^5].
The Church has put its weight behind a variety of unhelpful teachings on
sexuality over the centuries[^6]. But two, contradictory, ones stand out as
particularly salient today.
One is **prudishness**. Sex is evil. Sex is the origin of all evil: the Devil
seduced Eve, and Eve seduced Adam, and that’s when it all went downhill. Sex
is naughty. Sex is bad. Remember when Jesus said, ‘whoever looks at a woman to
lust for her has already committed adultery with her in his heart’ (Matt 5:28)?
He was saying that **all sexual desire** is sinful. Remember when Paul advised,
‘It is good for a man not to touch a woman’ (1 Cor 7:1)? He was saying that it’s
**bad** for a man to touch a woman **under any circumstances**.
This is, of course, false and dreadful teaching. The truth is that sex is a gift
from God, given so that we can bear his image by loving each other in this most
intimate and wonderful way. This is the consistent message of both Testaments.
But the idea that there’s somehow something **inherently wrong** with sex has
undeniably been a part of the Church’s teaching since at least the Victorian
period (and possibly a great deal longer than that).
The other is that sex is an **essential aspect of humanity**. If you aren’t
having sex, you’re missing out on an irreplacable part of your created purpose.
This has reacted explosively with the ideas of the Sexual Revolution. The
net result is that we have not so much been **freed to have sex** as we’ve been
**enslaved to have sex**. Virginity is an embarassment – both for men and for
women.
The radical Biblical idea that you can have a completely fulfilled life, deeply
enriched by loving relationships, **without** having sex or getting married,
is forgotten.
Everyone has been harmed by these teachings. But those who experience
significant attraction towards the same sex have been harmed double.
Faced by the impossible demands of bad Church teaching on sexuality on top
of the complexities of living with same-sex desire has left those people with
nowhere to go. They can _just say ‘no’_, and be made to feel that they’re
missing out on completing their full humanity. Or they can indulge their
same-sex attraction, and be judged not only promiscuous but a pervert to boot.
No wonder so many gay people have given up on the Church (and I haven’t even
talked about discrimination or violence towards gay people).
And no wonder that The Doctor has become strikingly anti-Christian at the same
time as it has become strikingly gay.
## What next?
Doctor Who’s criticisms of the Church hurt when they hit the mark. They hurt
more when they’re unfair. But that’s not the point. That point is this: the
Church has lost control of the conversation. And we’ve lost control of the
conversation because we threw it away. We threw it away encased in bomb-proof
concrete and left it to sink to the bottom of the deepest available ocean
trench.
If we in the Church are feeling hurt, we should start by feeling hurt by
our own sin.
The work of regaining trust on the question of sexuality will be the work of
decades: and that’s if we start working full-pelt right now. But there is hope.
I was encouraged last December by the visit to my local church, Bruntsfield
Evangelical, of _Living Out_[^7], a charity dedicated to helping churches across
the UK talk about sexuality. Ed and Andy, both same-sex attracted men, led us
through talks and discussions, and played recordings of perspectives from their
same-sex attracted female colleagues who couldn’t make it on the day.
They were primarily there to talk to us about how to support people like them:
same-sex attracted Christians who might be in our church who believe they are
called to singleness, helping them to thrive, living their true selves openly,
surrounded by love and empowered to share their love with others, just as we
want for everyone in our church. In fact, I heard some of the best news I’d ever
heard about my own sexuality, even though I am in the minority of people who
have never experienced significant same-sex attraction.
Whether or not you agree with their stance that the right place for sex is
within a marriage between a man and a woman, charities like _Living Out_ are
driving the Church in the right direction: serious, Biblical sexual doctrine
which helps us to realise in practice how we are all made in the image of the
God of love.
And how needed! How desperately needed! And Doctor Who gives us a little glimpse
of that, too.
Because not only is the series apparently anti-Christian, it also expresses a
need for Christian salvation. The Resurrection story is one which everyone needs
to hear, to have their death transformed to life, to have their full humanity
affirmed and celebrated and tended and to delight in it and to see it flourish.
To know the God of love and life, behind all and over all, with a plan and a
means to defeat the grip of death on the world.
So let’s keep having those frank conversations about sexuality in the Church.
It matters for all of us.
[^1]: Unless you’re bothered by
[plotholes big enough to fly a TARDIS through](https://www.doctorwhotv.co.uk/doctor-who-opinion-empire-death-2-102716.htm).
Oh, and you can probably skip episodes 1 and 3.
[^2]: For the record, this is flatly false. Though exceptions are widespread, the
overwhelming mainstream opinion is that priests should not bear arms. This is
not new, but has been the consistent orthodoxy, taken straight from Jesus’
teachings by the Church Fathers and maintained constantly since then. Can the
Church be legitimately criticised for its use of violence? Absolutely, let’s
have that discussion. Has the Church almost always been a **literal army**? No.
The Church has **never** been an army.
[^3]: See _God is Not Great_ by Christopher Hitchens, Chapter 11 for an epitome
example of this put forward by a prominent atheist apologist.
[^4]: In the spirit of charity, I have to accept that this admits of multiple
interpretations. It is possible that by calling them ‘creatures from hell’,
the priest is referring precisely to their unrepentant murderousness. However,
the priest has already tried to avoid the creature’s gaze by the pitiably
schoolboyish ploy of **looking elsewhere**, priming the viewers to think
that the priest is a silly coward: certainly not the kind to make a noble,
principled defence of justice with his head in the jaws of death. Plus, if
we interpret ‘creatures from hell’ as meaning ‘you don’t look like us so you
must be evil’, that would fit with the consistent framing of the series: that
religious morality is equivalent to bigotry.
[^5]: Though it must be remembered that many gay people have remained and
thrived within the Church in spite of the challenges, serving in diverse ways,
not least the priesthood.
[^6]: I found the Ezra Institute to give a good
[introduction to the history of Church teachings on sexuality](https://www.ezrainstitute.com/resource-library/articles/sex-and-the-history-of-christianity).
[^7]: For a great and humane introduction to what these guys are like, have a
listen to this [Living Out podcast episode](https://www.livingout.org/resources/podcasts/68/misstep-7-godliness-is-heterosexuality-the-plausibility-problem-7).
I get that if you haven’t grappled with issues like these before, it can be
really challenging, and you might find it hard to trust people who take a
different view to yourself. I know: I’ve been there myself! The best way to
start is to listen to a human voice.

201
scripts/blog/does_resurrection_ground_works.md Normal file
View File

@@ -0,0 +1,201 @@
---
title: Does resurrection doctrine give us unique reasons to work for justice?
description: >-
Tom Wright claims that the only Christian grounds for striving for justice now
is in resurrection doctrine. I'm not convinced.
pubDate: 2025-05-04
---
I've been reading Tom Wright's _Surprised By Hope_, defending his orthodox view
on resurrection. One of his key claims is that only by accepting the orthodox
position on resurrection can Christians justify striving for justice on earth.
To argue this, he needs to first show that resurrection doctrine does justify
striving for justice on earth, and secondly that the available alternatives fail
to do so. Firstly, the positive argument.
## Does resurrection give us reasons to work for justice?
Wright's argument depends on his view on what God's ultimate future will look
like: the present creation will not be abandoned, destroyed, or replaced, but
physically transformed into the new creation.
He argues that our work now has value, because, at the time when God transforms
the old world into the new, he will incorporate the outcomes of our good works
into the new creation, like an architect incorporating the works of many
stonemasons into a great cathedral.
He argues this on the basis of 1 Cor 15. I found it a struggle to find a good
justification for Wright's view in chapter 15 alone, but I did find some crucial
context in chapter 3, just before he begins the first of his many exhortations
to the Corinthians. It would be best to read the whole chapter, but here is
verses 11-15:
> For no one can lay any foundation other than the one already laid, which is
> Jesus Christ. If anyone builds on this foundation using gold, silver, costly
> stones, wood, hay or straw, their work will be shown for what it is, because
> the Day will bring it to light. It will be revealed with fire, and the fire
> will test the quality of each person’s work. If what has been built survives,
> the builder will receive a reward. If it is burned up, the builder will suffer
> loss but yet will be saved—even though only as one escaping through the
> flames.
This at least implies that the outcome of all our work will persist at least
until the time of judgement, when it will undergo testing, and those that pass
the test will enjoy a reward. You could read this as being like a quality check,
with God dishing out benefits to those that pass his assessment. But you could
read it in another way, more favourable to Wright. The works will be proven, and
the ones that withstand the process will themselves generate a benefit. You
could think of it like baking: when you put a cake in the oven, you prove
whether or not you got the recipe right; if not, it goes in the bin, and if you
did get it right, you get to enjoy the cake. So I agree that 1 Cor provides a
reason to think that the outcome of our works will somehow persist until the
time of judgement.
However, Wright doesn't just think that our works will persist until the time of
judgement, he also believes that they will at that time be transformed and then
incorporated into the new creation. 1 Cor doesn't directly justify this view. It
is, at least, coherent. Baking a cake in the oven transforms the dough.
So it seems reasonable to me to use 1 Cor to justify Wright's view that our
works will be transformed into the new Creation. However, that's before
considering any counter-arguments, and I have two which concern me.
One is that, as Ecclesiastes points out, the profits of our work will be laid to
waste by time. We don't know when God is going to bring about the new creation.
It could be tomorrow, and we should be behaving today in light of that
possibility - but of all the available possibilities, most of them are in the
distant future, so we should expect on average that there will be a long
interval between my deeds today and the judgement of them. Since time lays waste
to all our endeavours, we should expect that the profits of our work will have
vanished long before the judgement. Therefore, it would be unreasonable to work
for justice in the hope that our work will persist until the time of judgement
in order to be incorporated into the new creation, as Wright argues we should
do.
This first problem is probably the most important, but it's vulnerable to some
counter-arguments, which rather miss the point. So I'm going to move on to my
second problem, which is more niche, but which I think holds stronger against
counter-arguments.
In 1845, the HMS Erebus and HMS Terror left Britain on an expedition to explore
the Northwest Passage. In July 1845, they were spotted by whalers in Baffin Bay.
None of the sailors were ever seen alive again by Europeans. Both ships became
locked into ice in September 1846, and probably never sailed again. In April
1848, the remaining crew began a desperate 250-mile march to the nearest
European outpost. It is almost certain that all of them died.
Now, suppose that one of those sailors, between 1846 and 1848, did something
good. Maybe a major act of heroism, or maybe a small, kind word. I think it's
reasonable to suppose that someone, at some time, did something good.
All the profits of that good act have now completely perished. They are not
recorded in the small records the sailors left behind. None of the sailors who
benefited lived to pass on the benefits. None survived to pass on the light of
justice to the next generation.
At the time of judgement, then, there will be no remnant of this good act for
God to test, transform and incorporate into his new creation.
Does this mean that the good act was wasted? Was it only worth a shot just in
case, despite all the odds, the sailors made it home again?
I think this is a cruel conclusion: but it seems to be where Wright must go.
If it is implausible that the profits of all our good works will make it through
to the time of judgement, then we have to accept that 1 Cor 3 is true only in
some metaphorical sense, not that our works will literally persist in order to
be tested. And if that is true, then the premise of Wright's argument is false:
our work will not necessarily be incorporated into the new creation, so that
cannot generate reasons for working for justice now.
I am not convinced by Wright's positive argument. I would like to consider
alternative interpretations of 1 Cor. This is what Wright goes on to do.
## Do alternative views give us reason to work for justice?
The first view which Wright considers is the gnostic view that resurrection is
just an afterlife in heaven. He argues that this does not generate reasons for
working for justice now, but his argument is really the converse of the argument
in favour of his own view, which I've already considered above, so I'll move on.
The other alternative Wright considers is what he calls 'evolutionary optimism'.
You might also call this 'progressivism'. By this he means the view that the new
creation will be made the Church gradually building upon its own works,
generation after generation, approaching and eventually achieving God's perfect
standard by its works. This is the Victorian optimism which is still a powerful
force in our politics: that history is building upon itself, and progressing
from barbarism to civilisation, from brutishness to beauty, from tragedy to
justice.
He argues that this view, too, does not give us reasons to work for justice now.
His argument is intriguing: if our work for justice is condemned to only ever be
partially successful, then we have no reason to do it. In fact, contra the
evolutionary optimist, no amount of hard work on our part will ever achieve
perfect justice, and therefore if bringing in the new creation is all about our
works, we have no reason to strive for justice.
I'm intrigued by the premise that if we know that our work will at best be
partially successful, then we have no reason to do it.
This isn't how we ordinarily think: typically, if I think attempting to go to
the gym twice a week is going to be partially successful, I would say that this
generates a reason for me to go to the gym.
But I sense there may be an interesting meta-ethical thesis here: perhaps what
we ordinarily call 'partial success' is in fact a mistake, papering over what is
in fact simply a failure.
There's an obvious error theory: we had to create the concept of the 'partial
success' in order to generate reasons in the world as it appears, the world as
described in Ecclesiastes as 'vanity', where the best-laid plans of mice and men
gang aft agley, and all our works crumble into the dust eventually. In order to
think we had reasons at all, we needed to invent the concept of the 'partial
success'.
But is there a good argument for the view that there is no such thing as a
partial success?
In 1915, the HMS Endurance, under the command of Captain Ernest Shackleton,
became stuck fast in Antarctic pack ice in the Weddell Sea. In the face of
extraordinary challenges, Shackleton vowed to bring all his sailors back to the
UK alive: and in one of the most famous exploits of Antarctic exploration, he
succeeded.
It would have been wrong, had Shackleton vowed only to bring back _most_ of his
sailors. We know that he would have been able to bring _all_ of them back,
because he in fact did so. To strive for less would have been negligent.
This is to illustrate a general principle: we ought to strive for the best that
we are able.
Grant that necessarily, no particular justice is inevitable. It follows that
necessarily, it is possible to prevent all injustice. Therefore, perfect justice
is achievable.
And yet we _know_ that we will not achieve perfect justice. It's way too hard.
It does seem that I've proven a contradiction: both that perfect justice is
possible and that it is impossible. I expect these are two different kinds of
modality. I'm not too bothered to carefully distinguish them, as long we agree
that these two things can both be true in some sense. By analogy, consider that,
if you can run a mile in so many seconds, you can run it in a second less; that,
by sorites, it follows that you can run a mile in a minute; and that you
obviously cannot run a mile in a minute: it is too hard. Perfect justice is
perhaps a little like this: it is achievable in the sense that it is physically
possible for us to achieve it, but unachievable in the sense that it's way too
hard.
So, since perfect justice is in some sense achievable, it follows that we ought
to strive for it. But since we know we will not achieve perfect justice, it
follows that we cannot have a reason to strive for it: we cannot genuinely
strive for what we know we cannot do. Therefore, if you reject that God will
work to transform our world of vanity into something fundamentally different
where perfect justice is not only genuinely achievable but actually realised,
then you will be stuck in this hopeless tension, where you both must bring about
perfect justice, and have no reason to do it, because you have no hope of
success.
Where from here? I would really like to find alternative interpretations of 1 Cor,
and weigh them up against Wright's interpretation. It may be that, whatever the
counter-arguments, Wright's view is the strongest available. It may not.

95
scripts/blog/easter.md Normal file
View File

@@ -0,0 +1,95 @@
---
title: Why Easter is the best week of the year
description: >-
Based on a talk given to my colleagues at
<a href="https://www.scottlogic.co.uk">Scott Logic</a> for Maundy
Thursday, 2024.
pubDate: 2024-03-30
---
As you might have noticed, it is Easter this week! So I'd like to take five or
five minutes of your time to share why I – and about two billion other humans
going about the place just now – think Easter is the best week of the year. And
it's got something to do with a special Christian ritual called Communion.
Communion, at its heart, is about as simple a ritual as you can get. You get
together with a bunch of other people. You share some bread, and you share some
wine.
And it’s because of this ritual that so many people regard Easter as the best
week of the year. I want to explain to you why that is, and more than that, I
want to convince you that Easter is the best week of the year for you, too!
If you’ve passed by _The Hub_ at the top of Johnstone Terrace here in
Edinburgh recently, you might have notice the banner which is draped over
the railings just now – reading, ‘RITUALS THAT UNITE US.’
Now, that might seem like an odd idea. But wouldn’t that be great, if we
actually had a ritual which could unite us? Because the world could surely do
with a bit more unity right now. The world seems so divided, and sometimes it
seems like there’s no hope for real unity.
We can see that in our politics. We’re divided about foreign policy, about
taxation policy, about trade policy, about environmental policy.
And the conflicts that we have in this country seem pretty trivial when we
remember the conflicts that are playing out in other parts of the world right
now. In Israel and Gaza. In Sudan. In Russia and Ukraine.
And there’s plenty of conflict happening on the small scale, too. Often it’s the
smallest-scale conflicts which hurt us the most deeply. Your landlord pushes you
around. That friend you trusted like no-one else in the world lets you down. The
partner or spouse you loved like no-one else in the world – you end up fighting.
It’s possible that you’re going to be reading this right now with a heavy heart
because of a broken relationship in their life. And doesn’t that hurt more than
anything else we know?
When the world is groaning so heavily under the weight of conflict, and some
banner on _The Hub_ tells us a ritual can unite us, that seems so out of
proportion to the scale of the problem, doesn’t it? What can a ritual do? A bit
of old superstition? An excuse to divide people, maybe – what can a ritual do
to unite us?
Well, two thousand years ago, a man had a meal with his friends. Together, they
shared a meal of bread and wine – which, in that time and place, was the most
ordinary meal imaginable.
And yet, in that most ordinary event imaginable, something was happening which
was totally unimaginable. As this man, Jesus, shared the elements of this meal,
he made some extraordinary statements about what he was doing: ‘take, eat, this
is my body’ – ‘drink this, all of you; this is my blood of the new covenant.’
(The word ‘covenant’ means a promise.)
He told them he wasn’t just giving them bread and wine, he was giving his body
and his blood, and a promise.
Before Jesus ate another meal, he was flogged and nailed to a cross. His blood
was spilt and his body broken, even to death.
And yet, that wasn’t the end of the Easter story. Because three days later,
mourners turned up at Jesus’ tomb to pay their respects, and found the tomb
empty, the stone rolled away. Then they became the first of crowds of
incredulous eyewitnesses to see Jesus, the same Jesus who was killed on a cross,
alive.
Some magic trick, right? But this matters a hell of a lot more than just some
magic trick. Because Jesus became the first person in history to prove that you
really can both have your cake and eat it. He gave his life, and lived! As a
result, we can have his life and our own. We can join with Jesus through the
ritual of Communion which he established, and thereby, through Jesus’ body, join
together with everyone else who takes part in that ritual, as one body. Then we
can start living our brand-new, full-fat, original-recipe life overflowing with
generosity where we too can both give our life to others and enjoy it ourselves.
Indeed, Jesus taught us and showed us that it’s precisely by giving our lives to
others that we get to truly live ourselves.
This is why, in spite of all the division which persists in the world today, two
billion people regard this week as the best week of the year. Two billion
people, from every nation on Earth, speaking thousands of languages, of every
age and culture and gender and race, who defy the divisions of this world to
insist on joining together as one body in Jesus.
Because Jesus’ new covenant, his promise to all of us, is that in an apparently
hopelessly divided world, there exists real hope for unity. And that’s why
Easter is the best week of the year.

259
scripts/blog/euhwc_toast_to_the_lasses_2024.md Normal file
View File

@@ -0,0 +1,259 @@
---
title: EUHWC Toast to the Lassies 2024
hidden: true
description: >-
At the <a href="https://www.euhwc.co.uk">EUHWC</a> Burns meet in
Ullapool last weekend, I had the last privilege of giving the Toast to
the Lassies. Particularly for the benefit of those who weren’t there,
here it is in full!
pubDate: 2024-01-29
---
Had Burns, instead of his sweet bonnie Jean,<br>
his skills poetical for to mature<br>
had any one of our club’s lassies seen<br>
he would forever have remained obscure.<br>
If he had nothing but this box of worms<br>
Scotia would have been poorer, that I’m sure.<br>
Now none of us can claim to be a Burns,<br>
I’m no poetic master, still, I’ll have a punt,<br>
though let’s be clear, I’ll do it on my terms.<br>
I’ve everywhere avoided being blunt -<br>
politeness matters more than any schema -<br>
but it is hard when Isla’s such a cunt.<br>
It was a challenge to produce a terza rima<br>
I could recite withouten snoring;<br>
you’ve been so stiff I thought youse had oedema.<br>
The bother is this year is you’ll all been boring:<br>
no drugs, no sex, no gossiping or lies,<br>
no rock and roll, and hardly any whoring.<br>
But hey well, rules is rules, I’ve had to try!<br>
At least it can’t be worse than the reply.<br>
I’ll start with Audrey, the club’s senior member,<br>
for if there’s something that I say which disconcerts her,<br>
it’s fine: the poor old girl, she won’t remember.<br>
She likes to let us think she’s a hard worker<br>
but we’re electing a third social sec…<br>
it’s pretty clear she’s just another shirker.<br>
This lady, half American, half Czech,<br>
for study, moved to Scotland for to do<br>
American history – really, what the heck?<br>
The club is so much louder thanks to you:<br>
impressive vocals for just five foot two.<br>
That woman, Willow, reggles is bespeckled<br>
with her sickle and her fishing tackle<br>
shackled by the shins while she is heckled;<br>
the way that Willow waddles maks me cackle<br>
like a speckled jackal getting tickles,<br>
worth a shekel in the tabernacle;<br>
I chuckle muckle at her love of pickles<br>
which she wiggles when she has the heart<br>
while work for the Committee’s fickle trickles.<br>
Her modus operandi: <em>you can’t rush art.</em><br>
Her reimbursements programme’s going great;<br>
any day now, she’ll maybe even start.<br>
She cannot walk without Audrey, her mate:<br>
I wonder when they’re going to consummate.<br>
Although they make them pretty tough in Peebles,<br>
the thought of actually going up a peak<br>
fills Shona Lewis with the heeble-jeebles.<br>
New car? We miss your beautiful antique!<br>
How long before this one’s also up a creek?<br>
Once there was a lass called Hannah Collier<br>
whom even hell below regarded nasty,<br>
deeply despised by all that dwells there.<br>
Dating’s proceeding slowly for our lassie;<br>
not far from giving up til she beguiles<br>
a hot Italian in Southsider: classy!<br>
At first, Michaelo seems to be all smiles<br>
till it transpires he’s one of Dante’s demons…<br>
I guess it’s back to posters of Harry Styles.<br>
One day you’ll get a decent boy, keep dreamin;<br>
somewhere there waits a handsome Mr Collier.<br>
Hopefully when she meets him she’ll no be steamin.<br>
Hannah, I’m not sure why you chose to maul your<br>
poor skeleton at Subway (she’s still tetchy)<br>
and then abandon what remains of all your<br>
dignity at Ryvoan with a Frenchie!<br>
I think he wishes that he never met ye.<br>
And has a quiet Felicia e’er been seen?<br>
The energy she has is frankly wild.<br>
I’ve never seen a hillwalker so keen!<br>
<i lang="de">Ssie ischt raschtlos und nie gelangweilt</i>.<br>
She eats raw oats with soggy protein powder:<br>
a camping pot has ne’er been worse defiled.<br>
She uses what her Maker has endowed her<br>
with: her recorder skills are off the charts;<br>
youse think I’m joking, but I wouldn’t doubt her!<br>
This lass of the land of the Rot-Gold-Schwarz<br>
will soon depart, though long we might beseech you<br>
to stay. Of course, you’ll break all of our hearts,<br>
but mine most of all. Any time, Felicia,<br>
Creag Meagaidh calls, I know routes up the rear<br>
dark and under-explored that I can teach you!<br>
I won’t deny I think it’s rather queer<br>
the things you do with chickpeas, but no matter.<br>
You’re keen, you’re quick, you’re cool, that much is clear.<br>
In fact, I think you’d make a damn good faffer:<br>
swoop down on distilleries like the Luftwaffe.<br>
And now we come to our girl Emily Topness!<br>
You’re keen for social sec. You’d suit the role<br>
because… I’m not quite sure, it’s embdy’s guess.<br>
We met your sister, and she was just as dull.<br>
No, please drone on about Icelandic soil!<br>
Poor Joe here down the front’s bored out his skull.<br>
And since I mentioned Joe – I hate to spoil<br>
it for you – but you’ve got the inferior Joe,<br>
by Jove, no joke, it’s Jock here’s got the style!<br>
Nah, write the boy a sonnet, get in the flow,<br>
Whatever you produce’ll beat by thrice<br>
your Masters thesis. What’d you got to show<br>
for months of hunting for the butterflies?<br>
‘There weren’t any.’ Oh, and have some sense,<br>
cos I’ve heard rumours – I assume they’re lies –<br>
you’ve called yourself the ‘poet in residence.’<br>
You know you can’t compete, drop the pretence.<br>
Tereza was our gear sec for last year.<br>
She helped herself to stuff: that’s factual.<br>
Now when she asks to loan a bit club gear<br>
we have to ask her to provide collateral.<br>
She picked up tin whistle pretty sharp!<br>
Which is to say, she’s not a natural.<br>
She’s nowhere happier than under tarp<br>
gazing up at the moon and stars alone<br>
somewhere distant and remote like Glen Tarff.<br>
Now what to say about Lucy Ma-the-soooon....<br>
she likes… to faff… mm hmmm… well, moving on!<br>
And now we come to Emilie the French.<br>
She seems to be nice on the trips we see her<br>
but my distrust of frogs will ne’er be quenched.<br>
Claims she’s a ‘pharmacist’? So she’s a dealer.<br>
Need some pills in a pinch? You call, she’s there<br>
at your door in her rally-approved four-wheeler.<br>
One question we have is, why are you here?<br>
Most folk are in uni, you’ve no refutin<br>
you were kicked out after second year!<br>
Now the Engineering grad, Sophia Newton.<br>
Your namesake, Isaac, was a man convicted,<br>
constructed calculus; but no computin,<br>
not even Isaac’s, could’ve e’er predicted<br>
you’d drop the Eng for creative writing!<br>
now that’s what I would call a self-inflicted<br>
inflection point! It must be quite enlightening,<br>
but that doesn’t excuse when you give us an earful.<br>
The blood boils in our veins, the rage heightening,<br>
and you’re an American, that makes me fearful.<br>
What’s your secret? You have us knackered!<br>
What are you on to always be so cheerful?<br>
Now we approach the topic of Merzbacher.<br>
Wait, she’s not here? Abandoned ship?<br>
She says she’s informatics: so she’s a hacker?<br>
She has strong views, she lets her anger rip.<br>
Poor George got an earful, full of future advice,<br>
but why hasn’t she been on another club trip?<br>
We’re cruel to focus on this list of vice;<br>
the fact remains: she’s headstrong and nice.<br>
On Skye, a lady gave her poles to Sasha,<br>
which was really nice - I mean just the best -<br>
but Sasha really didn’t have to flash her.<br>
Quick history lesson: way back, RBS<br>
led the banking system to self-destruct<br>
and left taxpayers to pick up the mess.<br>
Since then, the name’s so irredeemably fucked<br>
they’ve had to ditch the brand once and for all.<br>
There’s one lassie who I need not instruct<br>
What, these days, the Royal Bank is called<br>
cos NatWest’s nasty history of scandal<br>
didn’t stop Booth from working there at all.<br>
Nothing motivates her more than to trample<br>
upon the working class. They set her free.<br>
She sank the pound quicker than the Belgrano,<br>
because ‘there is no such thing as society,’<br>
that’s how it is, is it? All right, I see.<br>
Now, coming all the way from Glenmore Lodge,<br>
it’s Ellie’s turn! We have done what we can,<br>
although I’m scared what she’ll put in my squash.<br>
She wasn’t into Benji, but listen man,<br>
you’re lucky that you dodged her drunken benders.<br>
You’ll wake up in a tent in Kyrgystan,<br>
as for how you got there, no-one remembers,<br>
and if you’d known you’d be sleeping next to Ellie,<br>
you would’ve brought some fucking ear defenders.<br>
She’ll wrap you in bubblewrap, from your ears to your belly,<br>
cotton clothes for none, and no complaining,<br>
applying safety to the max, spare socks in your wellies.<br>
She’s always at her Mountain Leader training,<br>
practicing her night nav in the locale,<br>
pursuing QMDs - unless it’s raining.<br>
But some water should not scare our gal!<br>
She’s had much experience with the wet as of late:<br>
after all, she got on well with our navy pal.<br>
What was the age of that particular first mate?<br>
Older than your ex - always part of the plan?<br>
Ah, of course! He was a spry twenty-eight!<br>
Youth’s for the losers, let’s get you a real man,<br>
mature and rugged, but kind and astute?<br>
Just make sure he’s not as old as your gran.<br>
One request we all have is you ditch the uke:<br>
never have strings been pluckèd quite so shitely;<br>
we would all much rather be hit by a nuke.<br>
And please shut up about your nice society.<br>
We are all glad you had a fun summer,<br>
but bringing it up throws us right back to sobriety.<br>
To lose you of course would be a bummer:<br>
that is, for your carefully groomed newcomers.<br>
Now time for the main woman, El Presidente!<br>
To here, it’s been like getting stones to bleed,<br>
but in Isla Burslem’s case we’ve material aplenty!<br>
As Holy Scripture says, ‘let those who lead<br>
well be worthy of double honour,’ so<br>
your bit is double length – it’s quite the screed!<br>
I’ll start off with her brilliant boyfriend – oh!<br>
Not boyfriend! Friend? To me this rather smacks<br>
of low commitment, but what do I know?<br>
So far, he’s disappointing, but on track.<br>
What’s he up to Isla: seven minutes? neat!<br>
Despite that, he is never holding back<br>
your blossoming romance with Dr Peat.<br>
Don’t deny it, that launch was pretty hard!<br>
It’s fifth base next: that’s photos of his feet.<br>
It’s fair to say her reputation’s marred.<br>
We all regret that we did once anoint<br>
her President: her premiership’s ill-starred.<br>
Hey - you’re meant to be in charge of this joint!<br>
You’re seldom seen cos of the mountaineering<br>
meets that you’re always on. You’d made your point<br>
before you chose to go off disappearing<br>
to <em>New Zealand</em>… we get the message! Plus<br>
we’ve had enough of all your domineering:<br>
maybe it’s time we put you on a bus!<br>
Nah, I’m just joking. All I’ve said’s refutable.<br>
But the boys, we mean this next bit, all of us,<br>
so stop me Isla if this isn’t suitable<br>
but honestly we think your mum is beautiful.<br>
Alas, I have to bring an end to this rhyme.<br>
I know it wasn’t much, in our defence,<br>
the fact you used ChatGPT’s a crime.<br>
I hope I’ve not caused over much offence<br>
don’t worry, that is it, I’ve said my bit,<br>
so I’ll turn from the ladies to the gents.<br>
Yeah, don’t look away now, we wrote this shit!<br>
I see you looking at your laces, Chris!<br>
Wit without real goodwill is not legit,<br>
so boys, don’t send sincerity to piss!<br>
Why did God say he’d take our hearts of stone<br>
and give us hearts of flesh? For this, for this!<br>
Here is flesh of our flesh, bone of our bone;<br>
love, and love nothing more but God alone.<br>

250
scripts/blog/god_is_not_great.md Normal file
View File

@@ -0,0 +1,250 @@
---
title: God Is Not Great, initial thoughts
description: >-
My spark notes on Hitchen's classic 2007 polemic against religion, plus some
initial thoughts on how I want to respond to it.
pubDate: 2024-04-14
---
These are my 'spark notes' on _God Is Not Great_, Christopher Hitchen's classic
2007 polemic against religion in all its forms, and call to adopt secular
humanism as its rightful replacement.
The book can be coherently read as a collection of independent essays. That
said, with a few exceptions, each chapter in _God Is Not Great_ contributes to
one of three main themes, and I think this is a helpful way of summarising the
overall movement of the book. The three themes I've identified are as follows:
<ol>
<li>
<a href="#1-religion-is-evil">Religion is evil</a>
</li>
<li>
<a href="#2-religion-is-untrue">Religion is untrue</a>
</li>
<li>
<a href="#3-secular-humanism-is-a-better-alternative-to-religion">Secular humanism is a better alternative to religion</a>
</li>
</ol>
As for the exceptions: Chapters 13 and 17 in part contribute to both the first
theme and the third, while Chapters 1, 3, 12 and 14 don't fit into these broad
themes, and are self-standing.
## 1. Religion is evil
- Chp 2
- Religion is violent, because:
- It has to be missionary, because it is insecure in its own beliefs (p17)
- It provokes tribalistic conflict, in a similar manner to racism (pp35-36)
- Chp 4
- Religion is bad for your health, because:
- Faith in medicine is a threat to religion's thrall (p47)
- Religious doctrines may contradict sound medical advice
- The right to religious freedom may be abused to deflect criticism of
unhealthy practices (p50)
- Religion has a special relationship with child abuse, and is incapable of
accepting open moral criticism of itself for this
- Chp 13
- Religion tends towards evil, because it requires fanaticism in order to
spread, and fanaticism tends towards evil (p192)
- Chp 16
- Religion causes child abuse, because:
- Terrifying children with eschatology is child abuse
- Religious education is propaganda and should not be inflicted on children
who are not yet mature enough to respond to it rationally
- Religion consistently mandates cruel genital mutilation of children
- Christians and Muslims have spread misinformation about masturbation,
which leads to dangerously sexually repressed adult men, which in turn
leads to sexual abuse of children by those men (as well as sexual abuse of
women)
- Child abuse in churches is not a case of a few bad eggs, it is
institutional and based on an ideological need to control the minds and
sexual organs of children
- Religion institutionalised torture in medieval Europe
- Religion makes honest and nuanced debate about abortion impossible, because:
- Nuanced debate is pushed out by extreme and implausible religious doctrine
- Religious people would rather use the unborn as objects of doctrine than
human beings in need of protection
- Chp 17
- Religion is the only reason anti-Semitism is possible
- Religion in its fullest expression is indistinguishable from
totalitarianism, because:
- The defining characteristic of both religion and totalitarianism is the
absolute right to rule of the ruler, even when they rule with caprice
- Religion and totalitarianism are also characterised by the need to
extinguish heresy with violence
- Religion and totalitarianism alike must propose a total solution to all
life's problems, require blind faith from its adherents, and demand all
aspects of life public and private be submitted to total supervision. This
doesn't bring out the best in us
- Religious/totalitarian systems are unable to take accountability and
therefore improve over time, in contrast to secular humanist systems
- History has proven this as fact:
- In the ancient world, religious totalitarianism was the normal form of
government
- Calvin, the inspiration for the Presbyterian Christian tradition, which
included South African apartheid, was the epitome of a totalitarian
dictator, demanding total control on the private lives of his citizens
in Geneva, on the pain of humiliation in this life and eternal torment
in the next
- History has also proven that, rather than standing in opposition to the
supposedly secular totalitarianisms of the twentieth century, religion
actually aided and abetted totalitarianism:
- Rome supported fascist movements throughout Europe, including Italy,
Hungary, Spain, and Ireland
- Rome accommodated Naziism by handing over control of its schools,
permitting the use of parish records to identify those with Jewish
ancestry, disbanding Catholic opposition political parties, declaring
Hitler's birthday a Church holiday, and running the 'rat line' to South
America after the military defeat of Naziism
- Although not quite as arse-licking as the Vatican, Germany's Protestant
churches also mostly capitulated to Nazi totalitarianism
- Japanese soldiers committed enormous atrocities across the Far East in
the name of their god-emperor, Hirohito
- The strategy of the Communists was first, to use religion as a prop to
gain power, and then to replace religion with itself. Notice the
striking commonalities between religion and communist totalitarianism:
- Infallible leaders
- A permanent war on heresy
- Institutionalised torture
- Scapegoating the innocent rather than accepting accountability for
failures
- Justifying any means necessary in order to achieve an ultimate end
## 2. Religion is untrue
- Chp 5
- Religion was a barbaric attempt to explain physical phenomena. Science now
does a much better job, so religion can be discarded as a redundant theory
- Chp 6
- Religion is 'solipsistic', which is to say:
- Religion divides the world into an in-group and an out-group
- When the in-group receives good fortune, that is interpreted as God's
blessing
- When the in-group receives bad fortune, this is inexplicable
- Whatever happens to the out-group is irrelevant
- This is a redundant theory which explains little, and therefore we should
not believe it
- Chp 7
- We should believe that the Pentateuch is a fiction, because:
- Exodus is inconsistent with the archaeological evidence
- Textual evidence in Deuteronomy suggests the texts were written much later
than the supposed events were supposed to have taken place
- Throughout the Pentateuch, Moses is referred to in the third person, which
is not consistent with the claim that Moses himself authored it
- The Pentateuch contains events we should _hope_ to be false, such as Moses
ordering multiple massacres, and the Ten Commandments classifying wives as
their husbands' property
- The Pentateuch - indeed, the whole Bible - is limited in scope to a small
corner of the Middle East, which is not consistent with the universal
nature of the supposed God who is supposed to have inspired its authorship
- Chp 8
- We should not believe the Gospels, because:
- Matthew and Luke disagree on the virgin conception
- Matthew and Luke disagree on the genealogy of Jesus
- Matthew and Luke disagree on when Mary and Jesus escaped to Egypt
- Luke dates the birth of Christ during both the reign of Herod in Judea
and the governorship of Quirinius in Syria, but these two events did not
overlap, so this is impossible
- As far as we know, the Romans did not, and would not, demand that the
people assemble in one place in order to be counted for a census
- The Gospels disagree about the Sermon on the Mount
- The Gospels disagree about the Anointing of Jesus
- The Gospels disagree about the Crucifixion and Resurrection of Jesus
- The Gospels disagree about the betrayal of Judas
- The Gospels disagree about the denial of Peter
- John suggests he thinks Jesus was probably born in Galilee to an
ordinary family with no proven link to King David's genealogy
- Mary's attested behaviour during the ministry of Jesus is not consistent
with the Nativity story
- John 8:3-11 (the stoning of the woman caught in adultery) is a later
insertion to John's Gospel
- Chp 9: the Koran is borrowed from a hotchpotch of Jewish and Christian
myths
- Chp 10
- Miracles (such as the Resurrection of Jesus) should not be believed,
because:
- See Hume's _On Miracles_
- Extraordinary claims require extraordinary evidence
- Believing miracles is comparable to believing reports of aliens (p144)
- Chp 11
- Religions are founded in credulity
- Chp 15
- The crucifixion story makes no sense, because:
- I'm supposed to be morally responsible for Adam's sin, but I amn't!
- Even Adam can't be fully blamed for Adam's sin, because he was set up!
- The Jews who crucified Jesus are supposed to be blameworthy for their
crime, even though the crucifixion was allegedly both necessary and
inevitable
- I'm allegedly given the free will to either accept or deny the offer,
even though denying the offer will lead to an eternity of torment: any
sensible God who cared a mite would not have given me the choice
- Religious rules are impossible to follow, and this leads to either
spiritual policing, organised hypocrisy, or both
## 3. Secular humanism is a better alternative to religion
- Chp 13
- Faith has inspired great heroism. But the heroism is better explained by the
heroes' humanism than it is by their faith itself
- For example, Martin Luther King Junior didn't really preach Christianity,
because he preached forgiveness, while Christ preached eternal torment for
the 'inattentive' (pp175-6)
- Chp 17
- Religion is the only thing sustaining anti-Semitism
- Secular humanist political systems can take accountability, respond to
criticism and improve over time, while religious systems exempt themselves
from legitimate criticism, stifling progress
- The only alternative to totalitarianism is pluralism, which is inherently
secular
- Chp 18
- Secular humanism has been a powerful positive force throughout history in
face of oppression by religion, as proved by example:
- Socrates proved that conscience is innate, and that a great way to mock
dogma is with satire which pretends to accept that dogma
- Lucretius, Democritus and Epicurus had better explanations for the way the
world worked than religion (which is why their work was suppressed in
Christian Europe and nearly lost forever as a result). Once rediscovered,
their ideas kick-started the Scientific Revolution in Europe
- Spinoza's deistic ideas had a huge influence despite Jews collaborating
with their Christian oppressors to try and ban his writing out of
existence
- Boyle and Voltaire may have been closet atheists, agnostics or deists
- Kant 'overthrew' the cosmological and ontological arguments for the
existence of God, and proved (by means of the Categorical Imperative) that
human decency does not require any theological assumptions
- Let's chuck some more names in the ring: Gibbon, Hume, Paine, Franklin,
Darwin, Einstein
- Jews were once doubly ghettoed: on the outside by oppressive Christians,
and on the inside by oppressive self-racialisation. Secular humanism freed
Jews from both these ghettoes, which in turn led to an outpouring of
secular Jewish brilliance
- Ancient Jews were on the road to a quasi-secular Hellenism, before that
was ruined forever by the tyrannical and fanatical Judas Maccabeus, with
disastrous consequences for the history of Western civilisation. The
Abrahamic religions we know today were not inevitable, and it is possible
to imagine what Western history would have been like without it
- Chp 19
- Secular humanism is a better alternative to religion, because:
- Religion requires clinging to immovable dogma and being unwilling to
change your mind
- Seeking truth requires being willing to change your mind
- Secular humanism is therefore on the side of seeking truth, and religion
is on the side of wilful ignorance
- Secular humanism is on the side of progress, because it is what enables the
expansion of scientific knowledge and the development of new technology
## The odds and ends
- Chp 1: an introduction to the themes of the book with little substantial
- Chp 3: titled 'A Short Digression on the Pig', it does what it says on the tin
- Chp 12: titled 'A Coda: How Religions End', it does what it says on the tin
- Chp 14: contrary to the hopes of some Westerners disillusioned with organised
Western religion, Eastern religions like Buddhism, Hinduism and Taoism are not
a solution to the problems of religion
I would be interested to try writing apologetic and/or evangelistic responses to
these ideas. But it's not a priority for me right now. Anyway, if I ever want to
come back to it, I'll probably come back to these spark notes to give me a head
start.

81
scripts/blog/llms_do_not_understand_anything.md Normal file
View File

@@ -0,0 +1,81 @@
---
title: LLMs do not understand anything
description: >-
Save this for the next time someone tells you that LLMs 'understand' things.
pubDate: 2024-06-13
---
LLMs do not understand what they are talking about. They just don't. It's not
how they're built. They have a statistical model of language, not a semantic
one. Philosophical puzzles about whether silicon can be conscious
_do not arise_, because LLMs are _not even close to the right shape_ for having
_anything like_ 'understanding'.
If you don't believe me, there are plenty of examples out there on the Internet,
but this is as good as any. It includes Chat GPT 4o explaining in detail why
\~\~A → \~A (which is a classical contradiction) is trivially true in classical
logic.
It's even better given that I had explicitly asked it to explain why that
sentence implies the trivial logic, not why it is trivially true. And even had
the explanation not been complete garbage from beginning to end, it would only
have shown that the sentence was _true_, not that it was trivial.
In other words, the output:
- Attempts to prove a contradiction (unprompted!)
- Confuses the concepts 'truth' and 'triviality'
- Is irrelevant to the prompt
In case you want it handy, I'll put the full conversation down below. The good
bit is at the end. The only edits I've made are to replace TeX syntax with
Unicode.
---
<dl>
<dt>
Me
</dt>
<dd>
You have the following axioms. <br>
1. |- (A -> B) -> A <br>
2. |- (~B -> ~A) -> (A -> B) <br>
3. |- (A -> (B -> C)) -> ((A -> B) -> (A -> C)) <br>
4. A -> B, A |- B <br>
Prove ~~A -> A. <br>
</dd>
<dt>
Chat GPT 4o
</dt>
<dd>
To prove ~~ A → A, we will use the given axioms and inference rules. Let's denote the axioms as follows:
1. ⊢ (A → B) → A
2. ⊢ (\~ B → \~ A) → (A → B)
3. ⊢ (A → (B → C)) → ((A → B) → (A → C))
4. If A → B and A, then ⊢ B
Here is the proof step-by-step:
1. **Axiom (2) Instantiation:**
From axiom 2, we have:
⊢ (\~ A → \~\~ A) → (\~\~ A → A)
Let B = A and A = \~ A.
2. **Intermediate Step:**
To use this axiom, we need to show \~ A → \~\~ A.
3. **Negation Introduction:**
By definition of negation, \~ A → \~\~ A is a tautology, because if A is false, then \~ A is true, and \~\~ A (which is A) must be false, making \~ A → \~\~ A always true.
4. **Modus Ponens Application:**
Since \~ A → \~\~ A is true, we apply modus ponens (axiom 4) to derive:
⊢ \~\~ A → A
Thus, we have proved \~\~ A → A using the given axioms and inference rules.

193
scripts/blog/my_feed_and_reading_list.md Normal file
View File

@@ -0,0 +1,193 @@
---
title: How I read things on the Internet now (no, of course I don't leave the terminal!)
description: >-
I've been sprucing up how I follow what's happening on the Internet. I can
now read all the things I want to read pretty efficiently, and enjoy doing
it, which is exciting.
pubDate: 2025-01-19
---
## The problem
I like to read things on the Internet sometimes. Or listen to things. Or watch
things.
Some interesting people write blog posts. Some beautiful people make music. Some
silly people make comedy sketches. Sometimes my friends are sharing stuff with
their friends, which means me (WIP on that front: I'll get back to that).
I want to know about it. I want to read interesting things that will educate
me. I want to hear beautiful music. I want to hear about what my friends are up
to.
I can't read everything as it comes in, it's too much: I'll need to sift a lot
of it out quickly. I'll need to sift through it regularly to stay on top of it:
like, several times a week, if not daily. And sometimes, while I'm doing my
daily sifting, I'll find something I really want to read, but I haven't got
time right now: I'll save this for the weekend.
For a while, I've been hosting [CommaFeed][commafeed] on [PikaPods][pikapods].
This has been OK ([OMG, RSS is cool][rss]), but the interface is just clunky
enough to make it a chore to use. That means I don't sift through stuff
regularly, and that means my feed piles hundreds of unread posts high.
It also hasn't been any good for distinguishing between stuff I don't want to
read *ever*, and stuff I don't want to read *right now* but will get round to
later. I could in theory use the bookmarking feature built-in to my browser,
but removing things after I've read them is too clunky, so I don't do it.
## The solution
Every morning, I open my terminal and run [newsboat][newsboat].
![newsboat showing how many unread posts I have at a glance in the opening view](./newsboat.webp)
I know the unread count is pretty fresh, because I've set up a systemd service
to run newsboat at startup to fetch the feeds.
I press `l` twice to open a post. Then I press `n` to navigate to the next
unread post until I run out of unread posts.
![newsboat displaying a post](./newsboat-post.webp)
If I encounter something I want to read later, but don't have time right now, I
press `b`, which runs a home-made bookmarking script. Here it is:
```bash
### readlater.sh ###
# newsboat passes a few arguments:
# the first one is the post URL
url="$1"
# I turn the URL into a suitable filename
filename=$( \
echo $url | \
sed "s/.*:\/\///" | \
sed "s/\//./g" | \
sed "s/\.html\$//g" \
)
filename="$filename".html
# If I've already got this in my reading list,
# I don't add it again
if [[ -a "$HOME/readlist/unread/$filename" ]]
then
exit 0
fi
# Otherwise, I download the post with curl
# and pipe it to a file in my reading list
# folder, ~/readlist/unread
curl "$url" > "$HOME/readlist/unread/$filename"
```
I can also run this script manually and pass it a URL of my choice at any time,
say, if I find an interesting article while browsing the Web.
When I want to read from my reading list, I run `readnow.sh`, which simply
opens my reading list folder, `~/readlist/unread`, in my terminal file browser
of choice: namely, [ranger][ranger].
![ranger showing the contents of my reading list with a preview](./ranger.webp)
Although ranger has a preview, I'll typically open the file up in my terminal
web browser of choice, which is [w3m][w3m] (plus a couple of custom
key-bindings). I've configured this to be my preferred web browser in ranger by
shifting it to the top of the list of HTML browsers in
`~/.config/ranger/rifle.conf`.
```conf
### ~/.config/ranger/rifle.conf ###
...
ext x?html?, has w3m, terminal = w3m "$@"
ext x?html?, ...
...
```
Having configured my default web browser in my ranger config, all
I need to do is press `l`.
![A post displaying in w3m](./w3m.webp)
No ads, no cookie popups, no giant banner images taking 2 seconds to load and
shifting the content all over the place: just the text I want to read. Isn't it
beautiful?
Once I'm finished reading the post, I'll press `q` to quit w3m and return to
ranger. Assuming I don't need to read it again, I'll press `dm` to move the
post to `~/readlist/read` - my way of marking a post as 'read'.
I've done this by writing a super simple script, `markread.sh`:
```bash
### markread.sh ###
filename="$(basename $1)"
if [[ -a "$HOME/readlist/unread/$filename" ]]
then
mv "$HOME/readlist/unread/$filename" "$HOME/readlist/read/$filename"
echo "Marked $filename as read."
else
echo "I couldn't find a file in ~/readlist/unread with the name: $filename"
fi
```
...and hooking it onto the custom keybinding, `dm`, in ranger:
```conf
### ~/.config/ranger/rc.conf ###
...
# map `dm` to run markread.sh in the shell, providing the active filename
# as the first and only argument
map dm shell markread.sh %f
...
```
## The result
I can now keep up to date, and I enjoy doing it.
I get not everyone likes to live in the terminal. I think the key takeaways
are:
- Make it really easy to sift through new posts
- When you sift, sift through every post, and for each one, either read it
straight away, or add it to your reading list
- Sift daily
- Make it really easy to add things to your reading list
- Make it really easy to browse, read things, and mark things read in your
reading list
- Set aside time to catch up on your reading list
- Make the whole thing joyful (both because joy is good, and because you won't
do it otherwise)
## Next steps
I still haven't really figured out social media. I'd like to stay up to date
with what my friends are doing, especially the ones I don't see very often. If
my friends are posting stuff on the Internet, I'd love to see it.
I still need to know:
- Are my friends posting stuff on the Internet?
- If so, where?
- What's the best way of subscribing to their posts - even if they live on
different websites and in different formats?
- What's the best way of sending and receiving comments/replies/reactions?
TBC. Answers on a postcard please.
[commafeed]: https://www.commafeed.com
[pikapods]: https://www.pikapods.com
[newsboat]: https://newsboat.org
[ranger]: https://ranger.github.io
[rss]: /blog/2024/05/02/no_more_youtube
[w3m]: https://w3m.sourceforge.net

BIN
scripts/blog/newsboat-post.webp Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 82 KiB

BIN
scripts/blog/newsboat.webp Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 63 KiB

58
scripts/blog/no_more_youtube.md Normal file
View File

@@ -0,0 +1,58 @@
---
title: How I made YouTube work for me
description: >-
I just learned YouTube channels have an RSS feed. This is terrific news.
pubDate: 2024-05-02
---
One of my bad habits in life is wandering through YouTube. I've always had
AutoPlay turned off, but I still found myself switching off and sleepwalking
from one 'recommendation' to the next.
A lot of what I watch in these times is crap. It's often when I'm tired and just
want to switch off, so naturally enough, longer, more emotionally or
intellectually material – in other words, exactly the kind of videos I actually
**do** want to watch – I ignore. (For me, it's often mediocre sketch comedy
videos. That's my poison, it turns out.)
Now, there is a time for switching off and doing something light. But I don't
want watching junk videos to be my answer. Any more than when I'm hungry after
a long day of work, I don't want to be resorting to junk food all the time. I
really want to get out of this bad habit, and though it's not a huge problem,
I've still not been able to fully kick it, either.
Still, there are YouTube channels I really _want_ to keep up with. Musicians.
Short film channels. Video essayists. And, yes, even one or two sketch channels:
though YouTube is a bit flooded with sketch comedy, some it is really good and
brings me joy, and that's a good thing!
This is where RSS comes in.
So if you don't know, an RSS file is a file someone puts on their website which
tells you, in a standard format, what pages are on their website. It's a pretty
old standard in Web terms, and it's very stable. There are a wide range of apps
out there, called 'feed readers', which you can use to keep track of RSS feeds,
notifying you when something new gets published in your website and putting it
in a feed for you to review.
It turns out YouTube channels have RSS feeds.
This is terrific news.
It means I can use my RSS feed reader to subscribe to YouTube channels, instead
of YouTube's own subscription system.
And that means I don't need any of YouTube's personalisation features. I can
delete all my search history, all my watch history, and stop them from
collecting any more. This means YouTube is now incapable of providing
recommendations that I'm actually likely to click on. Which is exactly what I
want.
This is what my YouTube homepage looks like right now:
![My YouTube homepage with a blank space where the recommendations used to be](/images/blog/2024/05/02/beautiful_youtube_homepage.jpg)
So I can both subscribe to the channels I'm genuinely interested in, and not
get sucked into watching junk.
Awesome!

46
scripts/blog/open_questions_about_sex.md Normal file
View File

@@ -0,0 +1,46 @@
---
title: Questions I have about sex
description: >-
Mark Vernon got me thinking about how the Church’s teaching on sex may be
evolving.
pubDate: 2024-12-17
updatedDate: 2024-12-17
---
I just listened back to [Mark Vernon][mark-vernon] talking about sexual desire
and Christian spirituality.
He recounts how his sex ed at an evangelical camp as a boy taught him to think
hard about holding hands with someone, just in case it led to something
‘immoral’. He felt he had been taught a sexual naïvety, which has perhaps
blinded people to the possibility of abuse – and misunderstood the gospel.
He puts forward an alternative to that naïvety, whereby sexual experience is an
image or a foretaste of love for, or union with, God, exploring thoughts by
Origen, Julian of Norwich, Dante, William Blake, Iris Murdoch, and various
parables and episodes from the Gospels themselves.
Part of what’s so interesting about this, is I feel I got my first proper sex
ed, as an adult, in an evangelical Christian context – and it pretty much lined
up completely with what he was putting forward.
I’m also currently reading (as is Mark Vernon) Diarmaid MacCulloch’s epic
history of Christian sexuality, [Lower Than the Angels][lower-than-the-angels].
But so far (I’m about a third of the way through and up to about the 5th
century) there have only been odd glimmers of positive Christian understandings
of sexuality. The overwhelming Christian consensus of the early Church
(according to MacCulloch) is that sexual desire is a symptom of humanity’s
fallen state, not our longing for God.
So I want to know: what changed between Mark’s experience and mine? How is
Christian teaching about sex changing right now? Was my experience typical of
other Christians growing up today, in evangelical churches, across the
denominational spectrum, across Britain, across the global Church? What _is_
the Christian consensus on sex now, and how old is it, and where did it come
from?
I guess I’ve got two-thirds of a gigantic tome to work through, first of all.
That’ll be a start. Any answers? Postcards please.
[lower-than-the-angels]: https://www.penguin.co.uk/books/313582/lower-than-the-angels-by-macculloch-diarmaid/9780241400937
[mark-vernon]: https://www.markvernon.com/the-gospel-sexual-desire-and-the-abuse-scandals-in-the-church-what-has-the-erotic-to-do-with-god

77
scripts/blog/paradox.md Normal file
View File

@@ -0,0 +1,77 @@
---
title: A paradox about 'should'
description: >-
I seem to have accidentally proven that drinking wine both is and is not a
thing you should do. Let's hope that wine doesn't disappear in a puff of
logical smoke.
pubDate: 2025-01-28
---
We're pretty familiar with the idea that there can be reasons for doing
something, and reasons against. Drinking wine is bad for your liver, but good
for your social life.
But look what happens if we express this in this way:
1. Drinking wine is bad for your liver.
2. You shouldn't do things which are bad for your liver.
3. All things you shouldn't do aren't things you should do.
4. Therefore, drinking wine isn't a thing you should do.
In contrast to this:
5. Drinking wine is a good social activity.
6. You should do things which are good social activities.
7. Therefore, drinking wine is a thing you should do.
Now both 1-4 and 5-7 seem like logically valid arguments with true premises,
but 4 and 7 are contradictory!
I don't think there's any use in complaining about premise 3. All that gives us
is the possibility that wine is both a thing you should do, and a thing you
shouldn't do. But that's an absurdity. Something can't be both obligatory and
forbidden at the same time. It's scarcely any better than a contradiction: it
is inconsistent with any useful concept of obligation.
Remember that we would quite like to know, at the end of all our argument,
whether we should drink wine or whether we shouldn't. 'Both' is not an adequate
answer, because it's not a useful guide for action: we can't both drink wine and
not drink wine. So if 'should' is to function as we need it to, 3 must be true.
I think a more profitable way forward is this. Let's re-write premise 2:
2. All things which are bad for your liver are things such that the fact that
that thing is bad for your liver is a reason not to do it.
We can similarly re-write premise 6:
6. All things which are good social activites are things such that the fact that
they are good social activities is a reason to do them.
Then premise 3, if it's to play the same logical role in the argument, would
have to read:
3. All things such that the fact that that thing is bad for your liver is a
reason not to do it is not a thing such that the fact that that thing is a
good social activity is a reason to do it.
Our re-written 2 and 6 seem to adequately capture the sense of the original,
but 3 is now obviously false. With our re-written sentences, we can avoid
generating a contradiction without doing any fatal damage to our concept of
'should'.
OK. Grant for the sake of argument that that was a good move. What have we
achieved? Have we actually solved the problem?
We started out with a pair of arguments which generate a contradiction. Our
move generated a new pair of arguments which don't generate a contradiction. So
what? Isn't the first contradiction still there?
Perhaps you could say that our re-written version of 2 (and so on) are more
accurate elaborations of the originals. Fine. But what about those originals,
then? You've still either got to say that they're true, false, or gibberish.
Producing your second argument hasn't convinced me to move my opinion about the
originals. The original premises seem just as true, and the original arguments
just as valid, as when we began.
What gives? Answers on a postcard as usual please.

BIN
scripts/blog/ranger.webp Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 76 KiB

74
scripts/blog/sapiens_on_religion.md Normal file
View File

@@ -0,0 +1,74 @@
---
title: Harari’s Sapiens on Religion
description: >-
In which I discuss why I think Harari’s characterisation of religion
is inadequate because it’s too materialistic.
pubDate: 2024-01-14
---
I’ve been slowly re-reading Yuval Noah Harari’s 2014 classic,
<a href="https://www.ynharari.com/book/sapiens-2">Sapiens</a>,
which apart from being ridiculously over-scoped and hilariously
under-evidenced, is proving delightfully entertaining.
I’ve just finished chapter 12, covering the world history of all
religion in thirty pages. Of course, at that level of brevity,
there will be many deficiencies. But here’s some thoughts - not
terribly well organised - which stand out to me.
Hurari generally assumes a materialist metaphysic (a problem which
blights the book more generally). Nothing exists except physical stuff.
This gives him severe tunnel vision. As a consequence of this
restricting metaphysic, he is forced to adopt limiting accounts of what
the role of religion is in world history, and therefore what religion is.
> The crucial historical role of religion has been to give superhuman
> legitimacy to [all social orders and hierarchies].
> Religion can thus be defined as <em>a system of human norms and
> values that is founded on a belief in a superhuman order</em>.
> <footer>p. 234</footer>
It might seem a little unfair to criticise Harari for giving a
materialist account of religion. <i>Sapiens</i> is, after all, a
materialist world history.
But this account is just one extreme example of how that project, to
give a materialist account of world history, will inevitably lack the
metaphysical resources to really understand the human story.
On Harari’s view, any human enterprise which attempts to understand
that which transcends direct human experience is at best an effort in
imaginative story-telling. All scientific theory, theology, ethics and
metaphysics either contorted out of all recognition into a pragmatic
fiction or is cast to the flames.
In particular, it’s a view which is incapable of taking seriously some
of the most important questions human beings have grappled with in the
course of their history. Those who know me won’t be surprised at which
ones I’m going to pick out: who was the being which made their covenant
with Abraham? How is that promise being fulfilled? And who the heck was
Jesus of Nazareth?
If Harari’s characterisation of religion is adequate - and the Abrahamic
faiths come under that banner - then those questions are reduced to
nothing more profound than Doctor Who fans arguing over ‘canon’. The
question of who God is becomes a mere tool for the organisation of
society, rather than a substantial and important question on a matter
of fact.
This is a shortcoming for its own sake: a materialist account of
religion cannot adequately account for the phenomenon of religion
itself.
But it is also a shortcoming even by its own lights. Without giving
serious consideration to the substantial matter of what Harari calls
‘religion’ (which, to his mind, includes the Abrahamic faiths,
Hinduism, paganism, animism, Buddhism, Shintoism, Confucianism,
capitalism, communism and Nazism), even the material facts are
inexplicable. Why would, as Harari is keen to point out, out, people
fight and die over and over again for a fiction?
The material facts themselves prove that ‘religion’ as he construes it
is not window dressing to the real story of history. It cannot merely
serve as a mechanism in the churning of material history. It is itself
the centre of the story.

139
scripts/blog/science_and_philosophy.md Normal file
View File

@@ -0,0 +1,139 @@
---
title: Why scientists need philosophers
description: This is a practice essay, OK? Don't shoot me if it's no good.
pubDate: 2025-01-24
---
I've recently been working at a
[Philosophy of Science MOOC]([phil-sci-coursera]) on Coursera, the online
courses website. Later on, they set you an exercise to write a short essay
addressing how philosophy can contribute to science.
So this is my punt. Life is too short to revise or, hell, even research this, so
don't take any of this as my settled opinion or as my best work. Whatever. With
that caveat in mind, it might still be entertaining at least, or maybe even
spark some thoughts.
---
Walk into a particular room in the Science Museum in Kensington, and you will
find yourself enveloped in a cavern of ironmongery and miscellanea. The shelves
droop under the weight of bolts, files, screws, bits, grinders, saws,
protractors, clamps and pins. The inventory scrolls endlessly past you, voices
reciting the lists of trinkets like an incantation.
This is the workshop of James Watt, meticulously reconstructed from the
original as it was left in his home in Birmingham after his death. This
entrancing space invites you to imagine a tireless creative, endlessly
tinkering away at his next contraption.
And yet for all that - and for all his immense valorisation as the lynchpin of
Britain's industrial revolution - Watt was reluctant to think of himself as the
engineer everyone else loved. He aspired to be remembered not as an engineer
but as a scientist.
To understand why James Watt, one of the most admired engineers who ever lived,
wished he were famed as a scientist instead, is to understand something
essential about modern Western thought.
In Watt's lifetime, scientists increasingly became the elite of society. In the
nineteenth and twentieth century, this trend only gathered pace. We constructed
our modern public health infrastructure on the advice of pathologists and made
medicine scientific. We funded scientific expeditions to map the world, even to
its remotest corners (in part to help us conquer it). We adopted radical new
economic policies in response to scientific economic theories. We built vast
infrastructure networks to communicate waves of invisible energy discovered by
pioneering physicists, and built nuclear plants to generate more of the stuff
by means of nuclear science. We even designed social programmes on the basis of
scientific anthropology. By the end of the twentieth century, scientists were
our prophets, priests and kings. Or so we thought.
In the 1990s, at the so-called 'end of history', it was assumed that there
would be no more need for social upheaval. Humanity had arrived at the ideal
system of social organisation. And among other ideologies - secularism,
libertarianism, democracy - an essential part of the package is that science
was our ultimate and incontestible way of securing knowledge about the world.
Yet now, that certainty is broken. Religious fundamentalism, whether Christian,
Muslim, Hindu, or else besides, is politically empowered in many places,
together with its rejection of science.
At the same time, the myth of science is ever more punctured. The supposedly
scientific West has increasingly come to appreciate that their scientific
heritage also includes much we would rather ignore - phrenology, race science,
systematic blindness to female bodies in medicine.
Meanwhile, scientists themselves are noticing that their holy calling has
turned out to be rather less holy than they had hoped. They find science
pulled between the competing demands of truth and tenure. Scientific knowledge
is locked behind the paywalls of exclusive journals, which even many academics
struggle to access, never mind the general public.
What then for science in the twenty-first century?
Yet there is another story available. It starts with confessing that the old
stories got things wrong in important ways. When we put ideology aside, science
has not been on an uncontested march to universal acceptance since Galileo.
There has been continual change, continual conflict, continual readjustment of
our ideas to the changing demands of the age.
In Galileo's day, it may have been a fight to establish that there was much to
be seen by simply looking. As empirical observation started to prove its worth
in the early modern period, thinkers wrestled with new problems: how to
reconcile the evidence of Scripture with the evidence of the senses? How to
understand how sensation can give us knowledge at all, granted that any
observer may be vulnerable to illusions, tricks and dreams? And if that's how
sense data work, what then for our mathematical or logical knowledge, which
seems to already bind the world even before we start experiencing it?
This centuries-long struggle culminated in the work of Kant, who in his 1786
magnum opus, the _Critique of Pure Reason_, set out a masterful - if infamously
obscure - system, which enabled thinkers to understand just how empirical
knowledge might work.
Yet two generations later, Charles Darwin lit the flame under new controversy
about the relationship between scientific and religious knowledge. His bizarre
and wildly imaginative theory of evolution by natural selection challenged the
Genesis Creation accounts, and this was soon to be followed up with the theory
of tectonic shift.
Some said that where scientists contradicted the authority of Scripture, the
word of God must always win. Others said that science alone had the keys to
knowledge, and if what the Bible said couldn't be proven scientifically, then
it couldn't be accepted. Some said that science and religion were two
incommensurable attempts to study the same subject matter, while others said
that they covered completely separate spheres.
Gradually, all of these views moved to the extremes. Now, most people (though
not all) agree that science and religion have overlapping spheres, and can
inform one another, but neither the Book of Nature nor the Book of Scripture
has the decisive final say.
Now, in our post-Christendom Western context, it's more important than ever to
understand how science and religion can talk to one another. Religious
minorities - as all religions now are in the West - are vulnerable to the risk
of becoming epistemic islands, cut off from the knowledge of the rest of the
community, unless we can find ways that science can talk across creedal
differences.
We need, too, for scientifically marginalised communities, such as non-white
people, whom science has ignored, or worse, to be more tightly integrated into
science, both so that knowledge might increase, and so that the benefits
knowledge gives might be fairly shared.
In light of these urgent needs, today's philosophers are considering science
not just as an epistemic problem, but as a social problem. As philosophers once
established science as the bedrock of modern knowledge, so philosophers today
have the task of figuring out how science can glue together our societies.
Science has been at its most dangerous when it hasn't been questioned. At all
times, as long as we practice science, we need to consider what it means, what
it means to do science well, how it can generate knowledge, and how it ought to
be used as a powerful instrument of change.
And perhaps that might justify James Watt in his obsession to be seen as a
scientist: since we can't get by just with practitioners. We need people who
can see our practices from the outside and shine a mirror back on us. If we
want science, then we need philosophers.
[phil-sci-coursera]: https://www.coursera.org/learn/philosophy-physical-sciences

194
scripts/blog/surprised_by_hope.md Normal file
View File

@@ -0,0 +1,194 @@
---
title: Surprised By Hope
description:
I've been working on my resurrection doctrine. Here's where I've got to.
pubDate: 2025-05-02
---
A couple of months ago, I was chatting to my friend Neil on the way home from
church, and in that conversation, I confessed to him that I had no idea what
happens to people after they die.
This might come as a surprise to some people who know me. Lots of people have
solid ideas about what happens to people after they die. For different people,
those ideas are very different. Here in Scotland, many people believe that death
is a final end. Many more believe that death marks a physical, and perhaps also
a spiritual, reunion with the rest of the universe, as the matter of your body
begins to be slowly digested and recycled: hence why ever more people are opting
to be cremated rather than buried in one piece. Other minorities believe in an
immortal soul that goes to some other place - be it heaven, hell, purgatory,
nirvana or reincarnation. I belong to the Christian community, which is supposed
to have clear answers on these questions passed down from ancient times, and
people who know me know that I think hard about doctrines. So it may be a
surprise that amongst all the convictions which people have all around me, and
amongst all my own convictions on other topics, I hadn't the faintest clue what
happens to people after they die.
If you are surprised, let me surprise you some more: I am of no fixed opinion on
a whole range of really important philosophical and theological topics, from the
existence of the soul to the purpose of sex, from the nature of the sacraments
to the metaphysics of the mind. But late last year, I set myself some New Year's
resolutions to address some of these questions. Not, by any means, to decide
once and for all the end of the matter: just to form a well-informed opinion.
Sometimes staying quiet isn't good enough: I'm aiming to rectify my silence on
these topics, because I think these topics are too important to ignore.
And one of the issues I picked out was this very issue: what happens to people
after they die? To that end, Neil recommended me a book by the conservative
Anglican theologian, Tom Wright, called _Surprised By Hope_, published in 2007,
at which point I was just learning to spell.
As a result of this book, I feel I understand what the Christian orthodoxy is,
and feel able to treat that view as my working assumption.
Wright defends traditional Christian orthodoxy. He claims that his view is
orthodox, and I'm roundly convinced that it is. When I wrote down in bullet
points what his view amounted to, I found that I had more or less re-written
half the Nicene Creed.
- Jesus Christ was crucified under Pontius Pilate.
- On the third day, he rose from the dead.
- He ascended into heaven and is seated at the right hand of the Father.
- He will come again in glory to judge the living and the dead.
- His kingdom will have no end.
- We look for the resurrection of the dead, and the life of the world to come.
Amen. That's pretty much it. All that I need to stress, to avoid under-stating
Wright's view, is that he specifically thinks that physical creation, including
our bodies, will be transformed into a new kind of physicality, including new
kinds of physical bodies for you and me, and that the 'world to come' means that
heaven and earth - which he regards as God's physical space and our physical
space - will be united.
He contrasts this orthodox view with several views common today amongst
Christians, some of which have even been muddled up with the traditional
orthodoxy:
- 'Jesus was raised to new life, spiritually, like a ghost.'
- 'Jesus literally ascended into the sky, as if he had an invisible jetpack: and
that's where he is now.'
- 'The Christian hope is that we will go to be united with God in heaven after
we die.'
- 'The Christian hope is that we will be snatched up to heaven at the rapture
and taken to a resurrection life there.'
- 'The Christian hope is that we will experience God's eternal life temporarily
before we die.'
- 'Jesus won't really judge anyone, because he loves everyone, and because he's
meek and lowly, not judgy.'
- 'The world will be redeemed through the work of the Church.'
- 'Only God can ever make a difference to the sinful state of the world, so the
only works we should care about now are "saving souls".'
I am happy to admit that I have often been guilty of most of these heresies. The
only ones I've never been tempted by are the 'rapture' view, and the thing about
the invisible jetpack.
Wright has not definitively put any of these ideas to rest for me. _Surprised By
Hope_ is just not that kind of book. It's not a treatise. It's actually quite
light on substantial argument in favour of Wright's position. Wright's main
achievement for me, isn't to convince me that he's right, but that his position
is a good starting point, a good place from which I should need to be convinced.
He does this chiefly by showing that his view is the consensus view of the New
Testament. (He claims to be showing it is the consensus view of 'the early
Church', but he never presents much evidence outside the New Testament, so I'm
being charitable by restricting his claim to the New Testament authors.) Say
what you like about Scriptural authority; if Mark, Matthew, Luke, John and Paul
all were convinced something was apostolic teaching, you'd better well take it
seriously.
If you want convincing, take a look for yourself. Some of the key New Testament
texts are John 5; Acts 17:30-32, 24:14-16; 1 Cor 15, 16:22; 2 Cor 4-5; Rom 6, 8;
Col 3:1-4; Eph 1:10; 1 Thess 4:14-18 and of course Rev 21-22.
You can also try convincing yourself that this is coherent with the Old
Testament hope, by looking at Isa 11, Dan 7, Ps 2, and having another look at
the assumptions behind Paul's behaviour in Acts 24:14-16.
The only significant problem texts I've found for Wright's view are 2 Cor 4-5
and Rev 21-22. In 2 Cor 4-5, Paul seems to plainly assert that we will have to
leave the body in order to face the judgement seat of Christ, and which makes no
apology for the assertion that, even though Christ has reconciled us to God, we
will still have to face judgement for our deeds - which seems to justify the
infamously un-Biblical doctrine of purgatory. If you assume that Paul's writings
express a completely consistent view, however, you will have routes out; in
particular, you could look at the language of Rom 6 and 8. Large chunks of
Paul's letter to the Romans also suggest, if taken out of context, that we will
have to leave our bodies behind, and that even those reconciled through Christ
will face judgement for their deeds - except that key verses contradict both of
those views. Clearly, that's not what Paul meant in Romans; so, you might argue,
it's not what he meant in 2 Cor either: provided you assume that Paul's writings
present a consistent view. (If Paul changed his mind, no explanation is
necessary why Rom and 2 Cor seem to be inconsistent: they could actually be
inconsistent in that case.)
Meanwhile, in Rev 21-22, John has a vision of a 'new heaven and a new earth, for
the first heaven and the first earth had passed away'. This directly contradicts
Wright's emphatic insistence that God's new creation will be continuous with the
first. For Wright, this isn't an academic detail, it's needed in order to give
us a motive to care for the world we've currently got. Without continuity, he
fears we'd be right to join those who are content with trashing the natural
environment because the whole thing's going to end up in fire and brimstone
anyway. Yet this piece of Revelation seems to permit exactly that.
If you were to defend Wright against Revelation, you might point out that
Revelation is a literal description of a vision John had, and is therefore not
in every detail an accurate picture of the future, but a metaphor, an image of
the future. (Fine, but if the wholesale replacement of heaven and earth is a
metaphor, what is it a metaphor for? If the literal future is continuity, why
not describe a vision of continuity?) And you may also assume that the entire
Bible is consistent on the matter of God's ultimate future, and on that
assumption, bring your analysis of the rest of the New Testament to bear.
Whatever you do with the problem texts, it seems clear to me that the
overwhelming weight of Biblical evidence favours the traditional orthodox
position over any of the alternatives. Given that, I'm happy to take it as a
starting point as I continue to think about what happens to people after they
die.
So, I may go back to Neil now, and say - maybe not quite yet 'I have an
opinion' - but at least 'I know what my working assumptions are.' I know what is
the orthodox Christian view: that is, the consensus view of the relevant
experts. The consensus view of relevant experts is generally a good place to
start.
I still have plenty of concerns, though. Here are my top three quandaries on
this topic now.
Firstly, it would be rather unsettling if the orthodox Christian vision for
God's ultimate future popped entirely into existence after the Ascension. The
apostles say that their teaching was given to them by the Holy Spirit - but are
we going to trust our entire doctrine on the future to what a small number of
men claim was told to them by an invisible being behind closed doors? If the
view of the New Testament authors is trustworthy, then it at the very least
needs to cohere very well with the Old Testament. The New Testament hope should
be woven deep into the Old Testament promises. I find Genesis, Daniel, Isaiah
and the Psalms promising, but I've only gotten skin-deep into comparing these
texts to the New Testament: I'd like to go both deeper into these texts, and
broader across the Old Testament.
Secondly, I want to hear the opposition in their own words. Wright very openly
admits that his view is currently a minority opinion even within Christianity,
despite being Christian orthodoxy. Given that is the case, it's reasonable to
expect the opposition to have some good arguments on their side. Wright has not
presented any strong arguments from opposing views, which makes me suspect not
that there are no good arguments, but that he has omitted to cover them in his
short and accessible book. And if there really are no strong arguments against
the traditional view, then we should expect powerful explanations as to why so
few people accept what apparently they should.
Thirdly, I have residual concerns from the metaphysics of mind. I recall from my
undergraduate days that continuity is a major concern amongst the relevant
experts. I think a minority of them even claim that the person who goes to sleep
and the person who wakes up again are completely distinct people who just so
happen to time-share the same body. If continuity is a major problem, then it is
a major problem for resurrection doctrine, too, which even in the New Testament
is compared to a kind of sleep, admitting that there is some kind of
discontinuity between the old body and the resurrection body. Add to this the
easily observable fact that many Christian bodies have rotted and are no longer
suitable for re-animation: their new bodies will have to be physically
discontinuous as well as mentally discontinuous with their old bodies. If I will
be given a new body, is it metaphysically plausible that the person who inhabits
that body will be the same 'me' that inhabits this body, now?
Much love all. As always, answers on a postcard please.

32
scripts/blog/tracking_pixels.md Normal file
View File

@@ -0,0 +1,32 @@
---
title: Tracking pixels
description: Concerning a notice in the privacy policy of an HR app
pubDate: 2024-04-10
---
I recently made the mistake of reading the privacy policy of the HR app my
employer uses. Take this choice excerpt:
> **Tracking Pixels**
>
> We may place tracking pixels in our Applicant Tracking System (ATS). These
> pixels are used for analytics purposes, including to track statistical
> information around when an email sent to you is opened.
Translation:
> We reserve the right to know whether you have read our emails, how long you
> spent reading them, at what time and in what physical location you read them,
> and any other information we can get our hands on. We reserve the right to do
> this as soon as you open an email, without asking for your permission first,
> and ask for your forgiveness later. That is, if you ever find out: which you
> won't, because we also reserve the right to hide this information in a Privacy
> Policy which we know you won't read, and we design our data-harvesting monster
> to be completely undetectable to anyone except determined experts, on purpose.
> We reserve the right to leave this data lying unused in a data warehouse where
> someone might steal it, and we also reserve the right to use it, if we can
> think of something satisfactorily dastardly to do with it. On the bright side,
> we probably won't, so having your data leaked to hackers and fraudsters is
> probably the worst you have to worry about. Unless we sell your data to
> advertisers, or a ~black hole of stolen data~ Large Language Model. Oh yeah,
> we probably reserve the right to do that, too. We doubt anyone will ever check.

BIN
scripts/blog/w3m.webp Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 91 KiB

48
scripts/blog/who_consecrates_the_temple.md Normal file
View File

@@ -0,0 +1,48 @@
---
title: Who consecrates the tabernacle? (Ex 29)
description: A quick, cursory and possibly completely rubbish observation I've made on Exodus 29.
pubDate: 2024-04-11
---
I've been reading Exodus recently, and the ending of Chapter 29 stuck out to me.
For pages and pages (since Chapter 25!) God has been giving Moses exact and
exorbitant instructions for how to build the tabernacle, what offerings to give
upon it, who the priests are going to be, what the priests are going to be
wearing. It's the fanciest IKEA manual you've ever read. No expense spared.
Everything is drowning in gold, silver, incense, myrrh, silk, fragrant oil, you
name it.
What's this about? The answer seems obvious: it's to sanctify the temple! To
make it beautiful enough and pure enough that even God could live there.
But right at the end of Chapter 29, in vv 43-44, there's this wonderful twist:
> ... and the place will be consecrated by **my glory**. So **I** will
> consecrate the tent of meeting and the altar and will consecrate Aaron and his
> sons to serve me as priests. (NIV, emphasis mine)
All this work, and at the end of the day, it's **God** who sanctifies his own
temple? So why all the fuss? Well, it looks like the final two verses of the
chapter give us the answer:
> Then I will dwell among the Israelites and be their God. They will know that I
> am the <span class="small-caps">Lord</span> their God, who brought them out of
> Egypt so that I might dwell among them. I am the
> <span class="small-caps">Lord</span> their God.
So on a cursory reading, it looks like God is telling them something
extraordinary about the kind of God he is. He is not like the other gods, who
demand this kind of worship to appease their pride. Nor is there, unlike the
other gods, anything the Israelites could do to make a place habitable for him.
That's because there's nothing the Israelites could do to cleanse themselves of
their sin: only the 'glory' of God could do that. And indeed, God has no need of
worship to puff up his pride, since his glory existed before the world was made,
and human beings can do nothing to either add to it or subtract from it.
Instead of a vain god who seeks tribute, this is a Father God who seeks to
'dwell' (literally to 'camp') among his people. The God who has put into motion
a rescue plan to pull the Israelites out of Egypt specifically for this purpose.
As it was in the beginning, when God dwelt in the Garden with Adam and Eve. This
is his mission. This is what he's like. And he wants his people to know this, so
that they can dwell together.

616
scripts/blog/word_hallucination_with_reference_to_llms.md Normal file
View File

@@ -0,0 +1,616 @@
---
title: What do academics think LLM ‘hallucination’ means?
description: >-
What does the word ‘hallucination’ actually mean in reference to LLMs when
used by academics? And where does that word come from? I’ve trawled the
academic literature for answers.
pubDate: 2024-07-16
updatedDate: 2024-07-17
---
The word, ‘hallucination’ as applied to LLMs has me enthralled right now. It’s
such a **weird** word. And problematic, I think.
If you don’t know what I’m on about, have a quick look at
[IBM’s introduction to the topic](https://www.ibm.com/topics/ai-hallucinations).
It gives you a good intuition what people are talking about, and helpfully
illustrates the deep conceptual confusions which have me gripped.
But this got me asking: what do people actually mean when they say that an LLM
is ‘hallucinating’? And why that word? Where does it come from, and how did it
end up in our mouths?
As far as I could see, nobody seems to have answered this question before, except
with the odd bit of unsubstantiated guessing. I think it’s important that we
know where this word came from and what it means.
So I ended up going on a bit of a treasure hunt. I’ve skimmed well over a
hundred academic articles to trace where the word came from, how it’s been used
over time, and how it’s spread from field to field.
The result is that I have what I hope is a pretty compelling story of where this
weird old word come from and what it’s supposed to mean. I’ve also got an
excruciating amount of evidence.
What I’ll do, is first present my story, and then step through the excruciating
detail, providing you with all my sources, so you can check my working, and
perhaps see what important evidence I might have missed. If you’re just here for
the headlines, you don’t have to read the excruciating bits.
Before I proceed, two caveats.
1. I am not a subject matter expert. Judge the quality of my arguments and my
evidence. Do not treat me as an authority.
2. For the purposes of this post, I’m just considering academic literature.
The ways that people use the word ‘hallucination’ in popular discourse may be
quite different. (For the record, I think they do: but that’s a story for
another day.)
## The long and short of it
So this is my theory.
In 1999, the word ‘hallucination’ was coined accidentally and off-hand by a
couple of researchers, Baker and Kanade, in the field of computer vision (CV).
The word caught on in CV, and came to have a fairly unambiguous meaning.
Given an image with missing pixels you want to fill in, ‘hallucinating’ meant
generating those missing pixels by means of information in a deep learning
model’s training data, rather than by means of any information contained in the
input image. (This was also known as ‘example-based in-painting’). It was a term
of art, not jargon.
Then, sometime between 2017 and the end of 2018, a few different researchers in
natural language processing (NLP) adopted the term. They were probably aware of
the term's usage in CV, and transferred it to mean something similar in NLP.
When these NLP researchers used it, they used it to refer to when a model
produces content which is irrelevant to the input, or contains information not
contained in the input.
When the word ‘hallucination’ first appeared in two or three papers in NLP, the
authors used it as a technical term for their own limited purposes. However, it
was soon adopted in various contexts for similar concepts.
For example, in image captioning, ‘hallucination’ came to mean producing a
caption which mentions an object which wasn’t depicted in the input image. In
machine translation, it came to mean generating a translation which was in the
right language, but which meant something completely irrelevant to the input.
In abstractive summarisation, it came to mean producing a summary containing
details which weren’t in the text the model was supposed to summarise.
When the term entered NLP, it underwent a striking transformation. It acquired
a **valence**.
In CV, the term referred to normal behaviour, and was value-neutral. It was just
a description of what a certain machine does. ‘We need to fill in these missing
pixels. The model will do that by hallucinating them.’
In contrast, in NLP, the term ‘hallucination’ was always used to describe
**abnormal** behaviour, and almost always used with a negative valency. In other
words, while CV folk used it to describe what they expected their models to do,
NLP folk used it to describe a **bug**.
(There were odd exceptions, though. You do get the periodic paper which insists
that there are positive applications of hallucination. But the fact they have to
insist so hard rather proves that that boat has already sailed. And even they
are still buying in to the idea that hallucination is abnormal behaviour.)
Now, that brings us up to about 2020. From 2020, we get another change.
From about 2020 until the present, academics have attempted to **jargonise**
the word ‘hallucination’. Broadly, they have so far **failed**.
When I say that people have attempted to ‘jargonise’ the word, I mean that
people are using it while assuming a specific definition. They are using it with
technical qualifiers like ‘extrinsic’. They are using it without explaining the
meaning or the context, as if you should already know what the word means. And
‘hallucination’ itself is becoming an object of study.
When I say that they have ‘failed’, I mean that there is no widely agreed and
specific meaning for the word, which in my mind, is a crucial feature of jargon.
However, the intended meanings tend to coalesce around the idea of an abnormal
behaviour of LLMs whereby they produce output which is not epistemically
supported by their input.
## Excruciata
OK, that was the long and short of it. Now for the excruciating detail. To
recap, I think these six claims are enough to support my theory:
1. Since 2022, academics have treated ‘hallucination’ as jargon
2. Since 2017/18, the word ‘hallucination’ has typically been used to describe
LLMs producing unfaithful output
3. Up to now, academics have not agreed a specific meaning for the word
‘hallucination’
4. The word ‘hallucination’ entered the NLP literature around 2017/2018,
probably from CV
5. The word ‘hallucination’ was not jargon in CV, but was a term of
art for example-based in-painting
6. The word ‘hallucination’ entered the CV literature in Baker & Kanade 1999
I’ll work through each of these in turn, showing how I got there from the
available evidence.
Feel free to use this as a cheap way to harvest citations for your own research.
### 1. Since 2022, academics have treated ‘hallucination’ as jargon
My evidence for this is twofold: surveys of ‘hallucination’ usually treat the
word as jargon, and many original research papers do, too.
First, let’s look at those surveys. These first five all belong together.
- Ji et al 2022 [^16]
- Liu et al 2023 [^30]
- Rawte et al 2023 [^42]
- Wang, Cunxiang et al 2023 [^48]
- Huang, Lei et al 2023 [^13]
All the latter four of these surveys lean heavily on Ji et al 2022 for their
definition of ‘hallucination’. Since they all source their definition of
‘hallucination’ from Ji et al 2022, it follows that their definitions are all
extremely similar, and often verbatim. Ji et al 2022 had this to say about the
word ‘hallucination’:
> Within the context of NLP, the most inclusive and standard definition of
> hallucination is the \[sic\] generation that is nonsensical or unfaithful to
> the provided source content.
They also made a distinction between **intrinsic** hallucination, which they
regarded as ‘the generation output that contradicts the source content’, and
**extrinsic** hallucination, which they regarded as ‘the generation output that
cannot be verify \[sic\] with the source content’. Some of the other surveys
pick up on this intrinsic/extrinsic distinction. To me, extending the word with
technical epithets suggests they regard the word itself as already having an
agreed, specific meaning.
Meanwhile, Li, Wei et al 2022 [^26], another survey, offers no definition
of ‘hallucination’, but uses it freely along with the intrinsic/extrinsic
distinction. This suggests they expect their expert readers to already know
some agreed meaning of the word ‘hallucination’. This is a hallmark of jargon in
my books.
Original research also shows the word ‘hallucination’ being used as jargon. For
example, Maynez et al 2020 [^35], despite being often cited when other authors
first introduce the term ‘hallucination’, provides no definition of the term.
They do, however, use the intrinsic/extrinsic distinction (this might be the
paper which coined the distinction).
All the following papers use the word ‘hallucination’, sometimes in passing
comments, sometimes as the principal focus of the paper, without attempting to
define the word.
- Mao et al 2020 [^32]
- Wang, Alex et al 2020 [^47]
- Lin et al 2022 [^29]
- Kumar et al 2022 [^21]
- Lee, Hwanhee et al 2022 [^23]
- Li, Junyi et al 2023 [^27]
- Guerreiro et al 2023 [^12]
- Dahl et al 2024 [^6]
- Song et al 2024 [^46]
Dahl et al 2024 also use the word without defining it, even as they tack on
their own bespoke technical epithets, ‘open-domain’ and ‘closed-domain’. Plus,
in many of these examples, the context is not enough to make clear what the word
‘hallucination’ is taken to mean.
All these examples demonstrate that the authors assume that their audience,
_viz_ technical experts, will come with an agreed and specific meaning of the
word ‘hallucination’ pre-baked and ready for precise academic application.
In other words, these authors, including both surveys and primary literature,
use the word ‘hallucination’ as if it is jargon.
### 2. Since 2017/18, the word ‘hallucination’ has typically been used to describe LLMs producing unfaithful output
Recall those surveys I referred to before. They mostly depended on Ji et al
2022’s definition, which was:
> \[...\] generation that is nonsensical or unfaithful to the provided source
> content. [^16]
Whatever we might think about ‘hallucination’, the word ‘unfaithful’ really
is a jargon word in the field of natural language processing, with an agreed,
specific meaning. An LLM is ‘unfaithful’ just in case it produces output which
is not epistemically supported by the input.
This is most often used in the case of summarisers. In that case, a summariser
is unfaithful just in case it produces a summary containing information which
is not implied by any information in the document it was supposed to summarise.
(‘Nonsensical’ is not a jargon term as far as I know. I have never encountered
any attempt in the NLP literature to define ‘nonsensical’, and although this
definition is often quoted, the term ‘nonsensical’ is rarely used in practice.)
Now, I am not claiming that this is the only way or even the dominant way in
which the term ‘hallucination’ has been used. Indeed, in section 3, I’m going to
show precisely that this is **not** the case: in fact, the word ‘hallucination’
continues to be used in diverse ways in the academic literature.
All I want to claim here is that it is **typical** for academics to use the word
‘hallucination’ to mean something in the rough area of unfaithfulness.
This is an important part of the story, because this is the meaning which links
the word most closely to its etymological roots in computer vision.
All I need to do really is provide enough citations. I know I haven’t read
everything. But if what was typical in my reading is unusual in the literature
at large, I’ve been the victim of some extraordinary bad luck. Have a look at
these papers:
- Durmus et al 2020 [^7] defines ‘hallucination’ as one of two kinds of failures
of faithfulness
- Huang, Luyang et al 2020 [^14] defines ‘hallucination’ as ‘creating content
not present in the input’
- Maynez et al 2020 [^35] introduces ‘hallucination’ as a cause of
unfaithfulness
- Zhao et al 2020 [^56] defines ‘hallucination’ as ‘including material that is
not supported by the original text’
- Nan et al 2021 [^38] defines ‘hallucination’ as putting out claims not
supported by the input
- Zhou, Chunting et al 2021 [^59] defines ‘hallucination’ as failures of
faithfulness in machine translation
- Mao et al 2020 [^32] clearly uses ‘hallucination’ to mean unfaithfulness
- King et al 2022 [^18] is aware of usages of ‘hallucination’ to refer to
unfactuality, but narrows in on failures of faithfulness, which they call
‘consistency’
So there you have it. ‘Hallucination’ has widely been taken to roughly mean
‘unfaithfulness’.
### 3. Up to now, academics have not agreed a specific meaning for the word ‘hallucination’
So we know that it was normal for academics to use ‘hallucination’ to mean
something like ‘unfaithfulness’. But was it also normal for academics to use it
for other meanings?
I think it was. There are a number of ways which authors used the word which are
not consistent with the ‘unfaithfulness’ interpretation.
The main way is that authors have assumed that in order to be a ‘hallucination’,
the output has to **contradict** something: either the input, or the training
data, or the facts, or itself. This contrasts with the ‘faithfulness’
interpretation, under which a hallucination can be perfectly consistent with
the input, the training data, the facts, and itself, as long as it isn’t
**supported** by the input.
All the following papers are clear that hallucination requires contradicting
something, usually the facts:
- Huang, Yichong et al 2021 [^15]
- Zhu et al 2021 [^60]
- Pagnoni et al 2021 [^40]
- Shuster et al 2021 [^45]
- Zhang, Yue et al 2023 [^55]
- Rawte et al 2023 [^42]
- Dahl et al 2024 [^6]
- Magesh et al 2024 [^31]
On the other end of the spectrum, a few authors seemed to believe that
hallucinated outputs **cannot** contradict the corresponding inputs. (This is
also inconsistent with the ‘unfaithfulness’ view.)
- Durmus et al 2020 [^7]
- Huang, Luyang et al 2020 [^14]
- Nan et al 2021 [^38]
There’s also disagreement on whether ‘nonsense’, or ‘incoherence’, counts as
hallucination.
On the one hand, Durmus et al 2020 [^7] argue that nonsensical outputs are not
hallucinations, since it doesn’t make sense to ask whether nonsensical outputs
are faithful. Likewise, Shuster et al 2021 [^45] contrast hallucination with
incoherence.
But on the other hand, Pagnoni et al 2021 [^40] include misleading and incorrect
grammar in their definition of ‘hallucination’, while Ji et al 2022 [^16]
explicitly include ‘nonsensical’ outputs in their widely-quoted definition of
‘hallucination’.
Some authors have managed to recognise the ambiguity of the term. Both King
et al 2022 [^18] and Farquhar et al 2024 [^9] acknowledge that the word
‘hallucination’ has been used variably to cover failures of faithfulness and
failures of factuality, and are careful to define their terms for their own
purposes to avoid being misunderstood in the context of this ambiguity.
So it seems that although many academics have treated the term ‘hallucination’
as jargon, in actual fact, there is no widely agreed specific meaning of the
word.
<hr>
I will slide in here with a quick side note. While some authors treated
‘hallucination’ as jargon for unfaithfulness or unfactuality, other authors
contemporaneously managed to talk about these topics without using the word
‘hallucination’ at all. Here are some examples:
- Cao, Ziqiang et al 2018 [^3]
- Li, Haoran et al 2018 [^25]
- Falke et al 2019 [^8]
- Goodrich et al 2019 [^11]
- Kryściński et al 2019 [^20]
- Cao, Meng et al 2020 [^4]
- Zhang, Yuhao et al 2019 [^54]
- Marcus & Davis 2020 [^33]
- Marcus 2020 [^34]
- Krishna et al 2021 [^19]
- Bai et al 2022 [^1]
- Weidinger et al 2022 [^50]
- Perez et al 2022 [^41]
- Min et al 2023 [^36]
- Muhlgay et al 2024 [^37]
Probably not a significant enough point to merit a whole section. But there you
are. I think it adds to the picture that ‘hallucination’ is failed jargon.
### 4. The word ‘hallucination’ entered the NLP literature around 2017/2018, probably from CV
So people have been using the word ‘hallucination’ in the NLP literature
recently. But where did it come from?
I think it entered the NLP literature somewhere around 2017/18, and probably
was borrowed from the field of computer vision (CV).
I can't offer a theory for a single, original usage of the word in NLP. But
there are three papers I feel are pretty close. These are the three oldest
papers I could find in the NLP literature which use the word ‘hallucination’.
- Wiseman et al 2017 [^51]
- Rohrbach et al 2018 [^44]
- Lee, Katherine et al 2018 [^22]
Both Rohrbach et al and Lee et al use a form of language which suggests they are
intentionally coining a technical term. Here’s Rohrbach et al:
> In Figure 1 we show an example where a competitive captioning model, Neural
> Baby Talk (NBT) (Lu et al., 2018), incorrectly generates the object “bench.”
> We refer to this issue as object _hallucination_.
And here’s Lee et al:
> These mistranslations are completely semantically incorrect and also
> grammatically viable. They are untethered from the input so we name them
> **‘hallucinations’**.
I have no reason to believe that these two papers are deliberately plagiarising
each other. It’s reasonable to assume that these two papers genuinely coined a
similar term for a similar phenomenon at the same time.
Wiseman et al 2017, in contrast, doesn’t look like it’s attempting to coin a
technical term. They just use the word once, presumably as a stylistic flourish
to help illustrate their point. But they do use it for a closely related
concept.
> \[…\] we see the model hallucinates factual statements, such as “in front of
> their home crowd,” which is presumably likely according to the language model,
> but ultimately incorrect (and not supported by anything in the box- or line-
> scores).
Just to add to the picture, Ehud Reiter, in his 2018 blog post [^43], gives us
an insight into the International Natural Language Generation conference of
2018. He claims that at that conference, ‘hallucination’ was a hot topic of
discussion, and cites Rohrbach 2018 to support his claim that ‘hallucination is
a well-known problem in neural approaches to image captioning’.
It might be fruitful to have a peruse of the
[INLG 2018 Proceedings](https://aclanthology.org/events/inlg-2018) to see
whether that interest in ‘hallucination’ was reflected in the written
contributions, or if it was mainly contained in verbal discussions. I haven’t
taken the liberty to do this myself: by all means, have a look yourself and let
me know what you find!
I think all this suggests that the word ‘hallucination’ entered the field
gradually, not with a bang, and perhaps spread by word of mouth at conferences
as much as it spread through published papers, at least at first.
But we still have to explain why all these different authors seem to have
independently come up with a similar meaning for the word ‘hallucination’.
I think the best explanation is that they got the word ‘hallucination’ from
computer vision (CV). If my fifth section is cogent, then ‘hallucination’ was
used in CV to refer to deep learning models generating data based on information
in their training data, not from information contained in or implied by the
input. This would explain how independent authors in NLP all independently
coined the word ‘hallucination’ in their own field to mean pretty similar
things. They were probably aware of the usage in the CV literature and adopted
it by analogy.
So, I reckon the word ‘hallucination’ probably entered the NLP literature from a
few authors independently, and they chose that word because they were borrowing
it from the CV literature, where it was already being used for a similar
concept.
But that depends on my next claim: so let’s look at that!
### 5. The word ‘hallucination’ was not jargon in CV, but was a term of art for example-based in-painting
I want to argue that ‘hallucination’, although it wasn’t jargon, was widely used
in the computer vision (CV) literature to refer to what was more technically
called ‘example-based in-painting’, that is, filling in gaps in images using
the information from training data baked into neural networks.
First look at Baker & Kanade 1999 [^2]. If my sixth section is correct, this
is where the term entered CV, but that’s not crucial to my argument here. Even
if I’ve missed some crucial evidence, it is nonetheless pretty certain that
they had a seminal effect on the use of the word in the field – they are early
in the field’s history, and widely cited by other CV papers which use the
word ‘hallucination’. They’re even cited as the origin of the term in machine
learning by the NLP paper, Farquhar et al 2024 [^9], a quarter of a century
later.
Baker & Kanade are writing about a new algorithm they’ve used in order
to increase the resolution of low-resolution images of human faces. The
intended application is for surveillance camera footage. They use the word
‘hallucination’ once in their title (‘Hallucinating faces’), once in their
abstract, and 79 times in the main body of the text.
Across those 79 occurrences, they are using it to refer to one of three things:
- Their algorithm: _eg_ their ‘face hallucination algorithm’
- The output of their algorithm: _eg_ ‘hallucinated faces’
- What their activity does: _eg_ ‘a face is hallucinated by our algorithm’
The closest they come to explaining what they **mean** by the word is in their
abstract:
> Although numerous resolution enhancement algorithms have been proposed in the
> literature, most of them are limited by the fact that they make weak, if any,
> assumptions about the scene. We propose an algorithm that can be used to learn
> a prior on the spatial distribution of the image gradient for frontal images
> of faces. We proceed to show how such a prior can be incorporated into a
> super-resolution algorithm to yield 4-8 fold improvements in resolution (16-64
> times as many pixels) using as few as 2-3 images. The additional pixels are,
> in effect, hallucinated.
(By ‘super-resolution’, they mean increasing the resolution of images. This is
typically reduced to a particular kind of ‘image in-painting’ problem, which in
general means filling in gaps in images.)
So, their idea is that, rather than limiting your algorithm to the information
contained in the input image, you can get better results by baking in
assumptions about the image to the algorithm.
Why is this supposed to work? They hope that for a clever enough algorithm,
the information that ‘this blurry splodge is a full-frontal portrait of a
human face’ will provide just enough information to accurately guess what a
higher-resolution version of the image would have been.
As for how that information, about what full-frontal portraits of human faces
look like, gets baked into the algorithm in practice: that’s done by training
a neural network on full-frontal images of faces. Hence why this field was so
close to NLP, which, in recent years at least, has leaned heavily on neural
networking or ‘deep learning’ approaches.
To recap, they don’t provide a specific meaning, but they do use it to roughly
mean the process of inventing missing pixels in images, not on the basis of
information contained in the input, but based on information contained in
training data instead.
To see how this term was picked up by later authors in the field working on
super-resolution, in-painting and other related tasks, see:
- Criminisi 2004 [^5]
- Fawzi et al 2016 [^61]
- Nazeri et al 2019 [^39]
- Xiong et al 2019 [^53]
- Xiang et al 2022 [^52]
But also notice other papers, on similar topics, which get on just fine without
using the term at all:
- Karras et al 2017 [^17]
- Zhou, Bolei et al 2017 [^58]
- Liao et al 2018 [^28]
And notice how Wang, Zhihao et al 2020 [^49] introduce the word, as an
alternative to another technical term:
> Face image super-resolution, a.k.a. face hallucination (FH), can often help
> other face-related tasks
The fact that nobody ever attempts to define the word ‘hallucination’, and it
doesn’t appear to be required lingo in any field, no matter how niche, to me
implies that it wasn’t treated as jargon in CV, at least during the period from
1999 until 2018ish. (The usage in reference to LLMs may well have gone back to
affect the usage in CV: I haven’t checked.)
So again. The word ‘hallucination’ was used in CV roughly to mean filling in
gaps in images by means of information contained in training data, baked into an
algorithm by training a neural network on many examples. It was used widely, but
not universally, and was not treated as jargon.
### 6. The word ‘hallucination’ entered the CV literature in Baker & Kanade 1999
I think that the word ‘hallucination’ entered the CV literature in Baker &
Kanade 1999. My argument for this is pretty straightforward.
I found the word ‘hallucination’ in Baker & Kanade 1999 [^2]. I couldn’t find it
in CV anywhere earlier.
To add to that, they don’t use the word ‘hallucination’ in a way which suggests
that they intentionally borrowed it from somewhere else.
My best theory is that they wanted a name for their algorithm to mark it out
from the competition, and the glove fit. I think that’s plausible enough, and
it’s consistent with the way Baker & Kanade use the word.
It’s also worth noting that according to Zhiwei et al 2009 [^57], the
foundational work in the field was published only in the same year (_viz_
Freeman & Pasztor 1999 [^10]). So there wasn’t really anywhere else for the word
to have come **from**. The only possibility is that they borrowed the word from
another field. If anybody reading this is able to suggest other fields which
used the word ‘hallucination’ before 1999, please let me know!
## Congratulations
If you’ve made it this far, you are a **hero**. Pat yourself on the back. Please
send me your corrections!
---
I have made little to no attempt to normalise these references into a standard
citation format. Please don’t tell the citation police.
[^1]: [Bai et al 2022. Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback. arXiv:2204.05862v1 \[cs.CL\] 12 Apr 2022](https://arxiv.org/abs/2204.05862)
[^2]: [Baker, Simon & Kanade, Takeo 1999. Hallucinating Faces. Tech. Report, CMU-RI-TR-99-32, Robotics Institute, Carnegie Mellon University, September, 1999](https://www.ri.cmu.edu/publications/hallucinating-faces-2)
[^3]: [Cao, Ziqiang et al 2018. Faithful to the Original: Fact Aware Neural Abstractive Summarization. The Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18)](https://cdn.aaai.org/ojs/11912/11912-13-15440-1-2-20201228.pdf)
[^4]: [Cao, Meng et al 2020. Factual Error Correction for Abstractive Summarization Models. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, pp 6251–6258, November 16–20, 2020](https://aclanthology.org/2020.emnlp-main.506.pdf)
[^5]: [Criminisi, Perez & Toyama, "Region filling and object removal by exemplar-based image inpainting," in IEEE Transactions on Image Processing, vol. 13, no. 9, pp 1200-1212, Sept 2004, doi: 10.1109/TIP.2004.833105](https://doi.org/10.1109/TIP.2004.833105)
[^6]: [Dahl et al 2024. Large Legal Fictions: Profiling Legal Hallucinations in Large Language Models. arXiv:2401.01301v1 \[cs.CL\] 2 Jan 2024](https://arxiv.org/abs/2401.01301v1)
[^7]: [Durmus et al 2020. FEQA: A Question Answering Evaluation Framework for Faithfulness Assessment in Abstractive Summarization. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp 5055–5070, July 5 - 10, 2020](https://aclanthology.org/2020.acl-main.454.pdf)
[^8]: [Falke et al 2019. Ranking Generated Summaries by Correctness: An Interesting but Challenging Application for Natural Language Inference. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pp 2214–2220, Florence, Italy, July 28 - August 2, 2019](https://aclanthology.org/P19-1213.pdf)
[^9]: [Farquhar, S., Kossen, J., Kuhn, L. et al 2024. Detecting hallucinations in large language models using semantic entropy. Nature 630, 625–630 (2024). 19 June 2024. https://doi.org/10.1038/s41586-024-07421-0](https://doi.org/10.1038/s41586-024-07421-0)
[^10]: [Freeman, W. T., & Pasztor, E. C. (1999). Learning low-level vision. Proceedings of the Seventh IEEE International Conference on Computer Vision. doi:10.1109/iccv.1999.790414](https://doi.org/10.1109/iccv.1999.790414)
[^11]: [Goodrich et al 2019. Assessing The Factual Accuracy of Generated Text. arXiv:1905.13322v1 \[cs.CL\] 30 May 2019](https://arxiv.org/abs/1905.13322v1)
[^12]: [Guerreiro et al 2023. Hallucinations in Large Multilingual Translation Models. arXiv:2303.16104v1 \[cs.CL\] 28 Mar 2023](https://arxiv.org/abs/2303.16104v1)
[^13]: [Huang, Lei et al 2023. A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions. arXiv:2311.05232v1 \[cs.CL\] 9 Nov 2023](https://arxiv.org/abs/2311.05232v1)
[^14]: [Huang, Luyang et al 2020. Knowledge Graph-Augmented Abstractive Summarization with Semantic-Driven Cloze Reward. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp 5094–5107, July 5-10, 2020](https://aclanthology.org/2020.acl-main.457.pdf)
[^15]: [Huang, Yichong et al 2021. The Factual Inconsistency Problem in Abstractive Text Summarization: A Survey. arXiv:2104.14839v1 \[cs.CL\] 30 Apr 2021](https://arxiv.org/abs/2104.14839v1)
[^16]: [Ji et al 2022. Survey of Hallucination in Natural Language Generation. arXiv:2202.03629v1 \[cs.CL\] 8 Feb 2022](https://arxiv.org/abs/2202.03629v1)
[^17]: [Karras et al 2017. PROGRESSIVE GROWING OF GANS FOR IMPROVED QUALITY, STABILITY, AND VARIATION. arXiv:1710.10196v3 \[cs.NE\] 26 Feb 2018](https://arxiv.org/abs/1710.10196v3)
[^18]: [King et al 2022. Don’t Say What You Don’t Know: Improving the Consistency of Abstractive Summarization by Constraining Beam Search. arXiv:2203.08436v1 \[cs.CL\] 16 Mar 2022](https://arxiv.org/abs/2203.08436v1)
[^19]: [Krishna, Roy & Iyyer 2021. Hurdles to Progress in Long-form Question Answering. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp 4940–4957, June 6–11, 2021](https://aclanthology.org/2021.naacl-main.393.pdf)
[^20]: [Kryściński et al 2019. Evaluating the Factual Consistency of Abstractive Text Summarization. arXiv:1910.12840v1 \[cs.CL\] 28 Oct 2019](https://arxiv.org/abs/1910.12840v1)
[^21]: [Kumar et al 2022. Language Generation Models Can Cause Harm: So What Can We Do About It? An Actionable Survey. arXiv:2210.07700v1 \[cs.CL\] 14 Oct 2022](https://arxiv.org/abs/2210.07700v1)
[^22]: [Lee, Katherine et al 2018. Hallucinations in Neural Machine Translation. Conference on Neural Information Processing Systems (NeurIPS 2018), Montréal, Canada](https://openreview.net/pdf?id=SJxTk3vB3m)
[^23]: [Lee, Hwanhee et al 2022. Factual Error Correction for Abstractive Summaries Using Entity Retrieval. Proceedings of the 2nd Workshop on Natural Language Generation, Evaluation, and Metrics (GEM), pp 439-444, December 7, 2022](https://aclanthology.org/2022.gem-1.41.pdf)
[^24]: [Lewis et al 2020. Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. 34th Conference on Neural Information Processing Systems (NeurIPS 2020)](https://proceedings.neurips.cc/paper/2020/file/6b493230205f780e1bc26945df7481e5-Paper.pdf)
[^25]: [Li, Haoran et al 2018. Ensure the Correctness of the Summary: Incorporate Entailment Knowledge into Abstractive Sentence Summarization. Proceedings of the 27th International Conference on Computational Linguistics, pp 1430–1441, August 20-26, 2018](https://aclanthology.org/C18-1121.pdf)
[^26]: [Li, Wei et al 2022. Faithfulness in Natural Language Generation: A Systematic Survey of Analysis, Evaluation and Optimization Methods. arXiv:2203.05227v1 \[cs.CL\] 10 Mar 2022](https://arxiv.org/abs/2203.05227v1)
[^27]: [Li, Junyi et al 2023. HaluEval: A Large-Scale Hallucination Evaluation Benchmark for Large Language Models. arXiv:2305.11747v3 \[cs.CL\] 23 Oct 2023](https://arxiv.org/abs/2305.11747v3)
[^28]: [Liao, L., Hu, R., Xiao, J., & Wang, Z. (2018). Edge-Aware Context Encoder for Image Inpainting. 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). doi:10.1109/icassp.2018.8462549](https://doi.org/10.1109/ICASSP.2018.8462549)
[^29]: [Lin, Hilton & Evans 2022. TruthfulQA: Measuring How Models Mimic Human Falsehoods. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics Volume 1: Long Papers, pp 3214-3252, May 22-27, 2022](https://aclanthology.org/2022.acl-long.229.pdf)
[^30]: [Liu et al 2023. TRUSTWORTHY LLMS: A SURVEY AND GUIDELINE FOR EVALUATING LARGE LANGUAGE MODELS’ ALIGNMENT. arXiv:2308.05374v1 \[cs.AI\] 21 Mar 2024](https://arxiv.org/abs/2308.05374v1)
[^31]: [Magesh et al 2024. Hallucination-Free? Assessing the Reliability of Leading AI Legal Research Tools. arXiv:2405.20362v1 \[cs.CL\] 30 May 2024](https://arxiv.org/abs/2405.20362v1)
[^32]: [Mao et al 2020. Constrained Abstractive Summarization: Preserving Factual Consistency with Constrained Generation. arXiv:2010.12723v1 \[cs.CL\] 24 Oct 2020](https://arxiv.org/abs/2010.12723v2)
[^33]: [Marcus & Davis 2020. GPT-3, Bloviator: OpenAI’s language generator has no idea what it’s talking about. Technology Review, August 22, 2020](https://www.technologyreview.com/2020/08/22/1007539/gpt3-openai-language-generator-artificial-intelligence-ai-opinion/)
[^34]: [Marcus 2020. The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence. Robust AI. 14 Feb 2020](https://arxiv.org/abs/2002.06177v1)
[^35]: [Maynez et al 2020. On Faithfulness and Factuality in Abstractive Summarization. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp 1906–1919, July 5 - 10, 2020](https://aclanthology.org/2020.acl-main.173.pdf)
[^36]: [Min et al 2023. FACTSCORE: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation. arXiv:2305.14251v1 \[cs.CL\] 23 May 2023](https://arxiv.org/abs/2305.14251v1)
[^37]: [Muhlgay et al 2024. Generating Benchmarks for Factuality Evaluation of Language Models. arXiv:2307.06908v2 \[cs.CL\] 4 Feb 2024](https://arxiv.org/abs/2307.06908v2)
[^38]: [Nan et al 2021. Improving Factual Consistency of Abstractive Summarization via Question Answering. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, pp 6881–6894, August 1–6, 2021](https://aclanthology.org/2021.acl-long.536.pdf)
[^39]: [Nazeri et al 2019. EdgeConnect: Structure Guided Image Inpainting using Edge Prediction. 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)](https://doi.org/10.1109/ICCVW.2019.00408)
[^40]: [Pagnoni et al 2021. Understanding Factuality in Abstractive Summarization with FRANK: A Benchmark for Factuality Metrics. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp 4812–4829, June 6–11, 2021](https://aclanthology.org/2021.naacl-main.383.pdf)
[^41]: [Perez et al 2022. Red Teaming Language Models with Language Models. arXiv:2202.03286v1 \[cs.CL\] 7 Feb 2022](https://arxiv.org/abs/2202.03286)
[^42]: [Rawte et al 2023. A Survey of Hallucination in “Large” Foundation Models. arXiv:2309.05922v1 \[cs.AI\] 12 Sep 2023](https://arxiv.org/abs/2309.05922v1)
[^43]: [Reiter 2018. Hallucination in Neural NLG, blog post, Nov 12, 2018](https://ehudreiter.com/2018/11/12/hallucination-in-neural-nlg)
[^44]: [Rohrbach et al 2018. Object Hallucination in Image Captioning. arXiv:1809.02156v1 \[cs.CL\] 6 Sep 2018](https://arxiv.org/abs/1809.02156v1)
[^45]: [Shuster et al 2021. Retrieval Augmentation Reduces Hallucination in Conversation. arXiv:2104.07567v1 \[cs.CL\] 15 Apr 2021](https://arxiv.org/abs/2104.07567v1)
[^46]: [Song et al 2024. FineSurE: Fine-grained Summarization Evaluation using LLMs. arXiv:2407.00908v1 \[cs.CL\] 1 Jul 2024](https://arxiv.org/abs/2407.00908v1)
[^47]: [Wang, Alex et al 2020. Asking and Answering Questions to Evaluate the Factual Consistency of Summaries. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pp 5008–5020, July 5-10, 2020](https://aclanthology.org/2020.acl-main.450.pdf)
[^48]: [Wang, Cunxiang et al 2023. Survey on Factuality in Large Language Models: Knowledge, Retrieval and Domain-Specificity. arXiv:2310.07521v1 \[cs.CL\] 11 Oct 2023](https://arxiv.org/abs/2310.07521v1)
[^49]: [Wang, Zhihao et al 2020. Deep Learning for Image Super-resolution: A Survey. arXiv:1902.06068v2 \[cs.CV\] 8 Feb 2020](https://arxiv.org/abs/1902.06068v2)
[^50]: [Weidinger et al 2022. Taxonomy of Risks posed by Language Models. FAccT ’22, June 21–24, 2022, Seoul, Republic of Korea. doi: 10.1145/3531146.3533088](https://dl.acm.org/doi/pdf/10.1145/3531146.3533088)
[^51]: [Wiseman, Hieber & Rush 2017. Challenges in Data-to-Document Generation. arXiv:1707.08052v1 \[cs.CL\] 25 Jul 2017](https://arxiv.org/abs/arXiv:1707.08052v1)
[^52]: [Xiang et al 2022. Deep learning for image inpainting: A survey. doi: 10.1016/j.patcog.2022.109046](https://doi.org/10.1016/j.patcog.2022.109046)
[^53]: [Xiong et al 2019. Foreground-aware Image Inpainting. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)](https://doi.org/10.1109/CVPR.2019.00599)
[^54]: [Zhang, Yuhao et al 2019. Optimizing the Factual Correctness of a Summary: A Study of Summarizing Radiology Reports. arXiv:1911.02541v1 \[cs.CL\] 6 Nov 2019](https://arxiv.org/abs/1911.02541v1)
[^55]: [Zhang, Yue et al 2023. Siren’s Song in the AI Ocean: A Survey on Hallucination in Large Language Models. arXiv:2309.01219v1 \[cs.CL\] 3 Sep 2023](https://arxiv.org/abs/2309.01219v1)
[^56]: [Zhao et al 2020. Reducing Quantity Hallucinations in Abstractive Summarization. Findings of the Association for Computational Linguistics: EMNLP 2020, pp 2237–2249, November 16 - 20, 2020](https://aclanthology.org/2020.findings-emnlp.203.pdf)
[^57]: [Zhiwei Xiong, Xiaoyan Sun, & Wu, F. (2009). Image hallucination with feature enhancement. 2009 IEEE Conference on Computer Vision and Pattern Recognition. doi:10.1109/cvpr.2009.5206630](https://doi.org/10.1109/cvpr.2009.5206630)
[^58]: [Zhou, Bolei et al 2017. Places: An Image Database for Deep Scene Understanding. arXiv:1610.02055v1 \[cs.CV\] 6 Oct 2016](https://arxiv.org/abs/1610.02055v1)
[^59]: [Zhou, Chunting et al 2021. Detecting Hallucinated Content in Conditional Neural Sequence Generation. Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pp 1393–1404, August 1–6, 2021](https://aclanthology.org/2021.findings-acl.120.pdf)
[^60]: [Zhu et al 2021. Enhancing Factual Consistency of Abstractive Summarization. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp 718–733, June 6–11, 2021](https://aclanthology.org/2021.naacl-main.58.pdf)
[^61]: [Fawzi et al 2016. Image inpainting through neural networks hallucinations. EPFL, Switzerland & IBM Research Watson, USA](https://www.cs.toronto.edu/~horst/cogrobo/papers/ivmsp2016.pdf)

47
scripts/migrateBlog.js Executable file
View File

@@ -0,0 +1,47 @@
const { glob, mkdir, readFile, writeFile } = require('node:fs/promises');
const readline = require('node:readline/promises');
const rl = readline.createInterface({
input: process.stdin,
output: process.stdout
});
async function main() {
await mkdir('blog-migrated', { recursive: true });
if (glob('blog-migrated/*').length) {
const answer = await rl.question(
'blog-migrated destination folder must be empty before starting, delete contents now?',
);
if (!/^[yY]([eE][sS])?$/.exec(answer)) {
process.exit(0);
}
}
const files = glob('./blog/*.md');
const writeOps = [];
for await (const file of files) {
const fileContent = (await readFile(file)).toString();
const [firstDelimiterMatch, secondDelimiterMatch] = [...fileContent.matchAll(/^---$/gm)];
const frontMatter = fileContent.slice(
firstDelimiterMatch.index + 4, secondDelimiterMatch.index
);
const content = fileContent.slice(0, firstDelimiterMatch.index).concat(
fileContent.slice(secondDelimiterMatch.index + 4),
);
const indentedContent = content.trim().replaceAll(/^/gm, ' ');
const yamlContent = `${frontMatter}\ncontent: |\n${indentedContent}\n`;
const filename = file.replaceAll(/^.*blog\//g, 'blog-migrated/').replaceAll(/.md$/g, '.yaml')
writeOps.push(writeFile(filename, yamlContent));
}
await Promise.all(writeOps);
}
main().then(() => {
console.log('Done!')
process.exit(0);
}).catch((err) => {
throw err;
});