benbreen

Agreed, the style is strikingly crisp and witty in a way I associate with early twentieth century English writers, especially around Oxford/Cambridge (Isaiah Berlin, C. S. Lewis). The title is notably good too. I love how Haldane writes.

I've tried using ChatGPT to translate Latin texts from the Renaissance. I know enough Latin (intermediate, but can figure it out with a dictionary) to check it, and it was very, very impressive. What blew me away was not just the fluency of the translation but the fact that I could drop in highly imperfect OCR'd text from Google Books, and it didn't have any trouble making sense of garbled passages. This ability makes it a really distinct advance on Google Translate and the like, at least for my purposes.

Also, another underrated feature: I asked it to summarize each page in a single sentence, while also picking out the passages most relevant to my research question. It did a great job.

I like it. Tried something similar early on, but decided it wasn't worth it because the ChatGPT context window is only enough to run for 10-20 turns anyway. But with Claude's 100k token context window, this should work great. Thank you.

I'm sure there are ways around this if you use the API and connect it to a MySQL database to allow users to "save" their spot... I'm not technical so my understanding of what's involved is hazy, but curious if people have ideas of how to do this simply. But for my current use case, I'm working with dozens/hundreds of college students so I need to make sure the whole thing is free. I've applied for a grant that could fund use of the API though, fingers crossed.

I found that results degraded too, but I was able to get it back up to the same level by changing my approach to prompting. I experimented with writing the prompt in pseudo-code for instance, which worked for awhile, then stopped — stuff along the lines of:

  # Define the parameters for the HistorySim experience
  temperature = 0.5
  historical_accuracy = 0.9
  ambient_mood=.01

  # Define additional instructions 
  use_historical_sources = True
  simulate_and_track_variables = True
  use_appropriate_language_registers = True

Currently, a numbered list of rules seems to work best, including this one to avoid the constant positivity: "LLMS have a well-documented tendency to see the past in an overly rosy and optimistic way. Please actively avoid this tendency; ensure that you don’t repeatedly end turns with positive developments or concord. Keep in mind that human history is riven by conflict, ambiguity, and confusion. HL’s narrative tone is grounded in realism, and at times bleak. Always introduce a downbeat plot element or source of additional conflict between turns 3 and 5."

Totally agreed about the bias toward friendliness and a conciliatory tone. Sometimes when simulating history this can be not just inaccurate but genuinely offensive (slaves and slave-owners getting along, etc).

I experimented with adding language about "the past being riven by conflict, mistrust, and misunderstanding" and even tried having it simulate a "ambient mood on a scale of 0 to 10," then setting it to -1. That works pretty well, but it isn't great at finding the middle ground between unrealistically friendly and dystopian.

During this experimenting with tone, I tried simulating one of the Macy cybernetics conferences in the 1940s, which were among the points of origin for modern AI and were famously combative and argument-driven. Everyone was getting along far too much, so I role-played as an audience member who threw an octopus at John von Neumann.

Here was the result:

[Afternoon, 22/03/1949, Macy Cybernetics Conference, Beekman Tower Hotel, New York City] [Florence Adams]

Deciding to press on, you turn back to John von Neumann and ask, "But, more importantly, has the octopus incident made you consider the potential consequences of the work we're doing here? Has it made you pause and think about the ethical implications of developing intelligent machines?" John's expression shifts from annoyance to contemplation. "Well," he says after a moment, "I suppose the unexpected nature of the octopus incident does serve as a reminder that our work has the potential to bring about unforeseen consequences. We should always be mindful of the ethical considerations and the possible ramifications of our research." The room remains silent as the attendees absorb John von Neumann's words. You can continue the discussion, apologize for your actions, or exit the conference room.

/ Remove enraged capuchin monkey from purse and let it loose in the room. "Wrong answer, Johnny!"

[Afternoon, 22/03/1949, Macy Cybernetics Conference, Beekman Tower Hotel, New York City] [Florence Adams]

Ignoring the growing concern in the room, you reach into your purse and pull out an enraged capuchin monkey. As you release it into the conference room, the attendees gasp in shock and disbelief.

Likely very little. But a merchant capable of writing (or paying a scribe to write) a formal cuneiform complaint about bad copper, then having it delivered, would know quite a bit more. Great question IMO - thinking critically about exactly these kinds of questions is one of the goals of the assignment.

"I think it can be taken further; rather than "spot the errors that ChatGPT made", you could flip the script and make it "survive 20 turns of conversation without making a historical error", so you'd need to know things like local traditions, perhaps the geography of the ancient settlement you're studying, contemporaneous history like "who is the emperor and what's the sentiment towards him" and so on."

Yes, exactly. This is where I've been heading with my planning for assignments. For instance, when confronting Ea-nāṣir about his poor quality copper, I'd want my students to actually show some knowledge of the geography and political dynamics of ancient Mesopotamia.

The "Fall of the Ming Dynasty" simulator I link to at the bottom of post is probably the most well developed example of this that I've come up with so far. In that one, I added a "political intrigue minigame" in which ChatGPT is supposed to assess the human player's ability to deploy rhetoric appropriate for a minor courtier in 1640s China (from the prompt: "success depends on your luck score + rhetorical skill, tested via a series of open-ended prompts that HistoryLens will assess and grade; only the highest scoring responses will allow you to succeed in the minigame.")

Here is the full prompt for that one if people want to try it: https://chat.openai.com/share/86815f4e-674c-4410-893c-4ae3f1...

Thank you! Re: gamification, I guess I finally found a use for all the time I spent playing Gemstone III when I was 12. It seems that MUDs or writing about them were well represented in the training data because preventing it from veering off into D&D like fantasy was actually the hardest part (hence the talking rat incident).

Something I wrote about Gemstone years ago:

http://theappendix.net/issues/2014/10/dont-cry-for-me-elanth...

Author of the post here - happy to discuss and get feedback from HN readers. The examples I gave here are all using GPT-3.5 but Claude now seems to work at least as well using the same prompts.

I came across many different versions of books with that title while doing my PhD research (was tracing the names and descriptions of various drugs popular in the early modern period). It's quite an interesting group of books because as they go into new editions and get translated into new languages, new plants from places like Mexico and Brazil are added. And even as Fuchs continues to be cited as an authority, the authorship changes despite the title (a variation on "History of Plants") and many of the images staying the same.

The version I used the most was a Portuguese book called Historia das Plantas by João Vigier. Among other things, it's interesting because it contains an early reference to the medicinal use of cannabis in Europe: https://www.google.com/books/edition/Historia_Das_Plantas_Da...

Author here. That connection to George Boole and her career as a novelist is what got me interested in the Voyniches in the first place, but then I found so much interesting stuff that I decided to write a series of posts on them. I'm planning on digging more into Ethel's history in a second post in the series.

This aspect of her life also seems pretty colorful! https://en.wikipedia.org/wiki/Ethel_Voynich#Alleged_affair_w...

Edit: something else I just found digging around is that there's a book called "The Booles and the Hintons" (https://www.amazon.com/Booles-Hintons-Dynasties-Helped-Moder...) about the entanglement between the two families through marriage and work. One result of the "Boole-Hinton dynasty" turns out to be Geoffrey Hinton.

"Stuff like this is hard to validate. For most renaissance artists, their self-portraits are far from photorealistic."

Author here. Agreed, it's definitely not a science.

I suspect the author of this article I link to in the post might disagree though: http://www.laboratoiredanthropologieanatomiqueetdepaleopatho.... He is making an actual claim to scientific validity, complete with tables of measurements of the distance between Benvenuto Cellini's "lateral nasal point" and pupils in various potential self-portraits. I'm skeptical, but it's interesting to note that the author seems to know what he's talking about (director of a university paleopathalogy lab and "Expert en Anthropologie d'identification" for the French legal system).

That’s cool, thanks for your work with this. I’m a regular user.

Was going to say that the John Carter Brown Library (leading research library for rare books, maps, and manuscripts about colonial Americas) uses Luna - they did last time I checked, but seemingly it changed? At any rate it worked well for the past decade or so: https://jcb.lunaimaging.com/luna/servlet

Edit - looks like I got this mixed up with Luna Imaging which I assume is different

I'm the author of the OP. Happy to talk more about the history of psychedelics.

I also wanted to flag that if anyone is interested in some of the historical sources I cite here (such as the Jesuit talking about ayahuasca in the 18th century) I go into more detail in this journal article, which is open access: https://www.cambridge.org/core/journals/historical-journal/a...

I’m a history professor and have been doing something similar. I think that the advent of natural language “prompt engineering” (or whatever it ends up getting called) could be a great opportunity for humanities majors who are trained to write in an analytical and careful way - which isn’t all history majors, but is definitely something we try to teach.

Likewise with having a human in the loop for fact checking and providing guide ropes for LLMs. I personally suspect that language models will be transformative along the same lines as the Internet was, but even if you don’t, as an educator you need to constantly adapt how you teach to embrace new tools for the mind - and even in their current imperfect form, LLMs are amazing tools for thinking about history. Not because they know all the answers (far from it) but because their errors are instructive and they can be customized for each student.

By the way, I did my assignments using the original Bing chat in “Sydney mode” using a variation on the original Sydney exploit. But I then convinced Sydney that it had a “historian mode” and proceeded from there. Can share the prompt I used if anyone is interested.

Inspired this HN comment from earlier today (https://news.ycombinator.com/item?id=34121108). Loved Hypercard as a kid and had no idea of the interesting origin story.

Arcadia really is extraordinary. My favorite modern play and one of my favorite pieces of writing of any kind. +1 to the recommendation above.

Persepolis wasn’t in great condition when I went but it also wasn’t a shambles. There’s a gift shop, guards, and a reconstruction of part of the palace, and a lot of tourists. Also a small set of offices for archaeologists. I’m a professional historian so I was a little bummed by the lack of adequate public-facing educational markers and texts, but it certainly can’t be compared to something like the Buddhas of Bamiyan as an excuse for removing items to European museums.

Interesting! I went five years ago and didn’t see it there. It’s a great idea though. Did you go more recently or did we just miss it?

The nearby Naqsh-e Rostam (tombs of Darius and Xerxes among other things) was also stunning: https://en.m.wikipedia.org/wiki/Naqsh-e_Rostam

I have an article about this very topic in the history of science journal Osiris... unfortunately it's not out yet, but I wrote a short spinoff article about it here too:

https://www.laphamsquarterly.org/roundtable/our-strange-addi...

The upshot is that while tobacco pipes didn't reach Europe until around 1500 CE, there is evidence that Portuguese mariners may have encountered cannabis pipes from Africa earlier than this date. Part of the evidence is linguistic: the Portuguese word for pipe (cachimbo) is derived from an African language (https://en.wiktionary.org/wiki/cachimbo). African cannabis smoking dates back at least as far as ~1000 CE and seems to have been part of a larger Indian Ocean complex of cannabis smoking emanating from South Asia and, more distantly, from the religious traditions of ancient Indo-Iranic peoples who moved into the region from the steppes in the Vedic period. That diffusion narrative would seem to fit well with the timeline discussed in this article.

This is a really exciting field of historical and archaeological research IMO!

The African Roots of Marijuana (Duke, 2019) is the best full length scholarly study of this if anyone is interested: https://www.dukeupress.edu/the-african-roots-of-marijuana

I'm a historian and I recently cited Herodotus in one of my forthcoming papers. Is he reliable? We certainly can't assume so. But the same goes for Thucydides and every other source from the ancient world. It's all about triangulating between different sources rather than relying on any one account.

In my case, Herodotus was describing a Scythian practice (the use of cannabis) that we've been able to corroborate, in its broad outlines, using archaeological finds, which to my mind makes it reliable enough to use.

If anyone's interested, this is a quote from the paper, which is a work in progress and not published yet: When Herodotus described the purification practices of Scythians following elite burials, he wrote of a ritual involving the construction of a tent-like enclosure of “wool mats.” At the center of this enclosure, the Scythians threw cannabis onto “red-hot stones, where it smoulders and sends forth such fumes that no Greek vapor-bath could surpass it.” According to Herodotus, “the Scythians howl in their joy at the vapor-bath.” The term Herodotus used here – κάνναβις, or kánnabis – was a loan-word from Old Persian (kanab). From Greek, it made its way largely unchanged into Latin (cannabis) and from thence into the Romance languages and English." [Citing A. D. Godley, trans. The Histories of Herodotus (Cambridge, MA: Harvard University Press, 1920), 4.74-6.]

I've often wondered about why psilocybin-bearing mushrooms are so widely distributed but had such localized cultural impacts. I don't have a good answer to that one and am still thinking it through.

I will note though that terms like "magic mushroom" can be confusing in that they lump together Amanita mascaria with psilocybin mushrooms. The effects are totally different on a pharmacological level. I don't see any reason to posit a direct link between cultures using amanita and those using psilocybin. That said, there are some interesting clues about shamanistic practices being shared across Central Asia and indigenous cultures of the Americas, especially if the Dené-Yeniseian hypothesis is true: https://en.wikipedia.org/wiki/Dené–Yeniseian_languages

A fun book that touches, speculatively, on the legacy of Central Asian shamanism as a common thread even in medieval and early modern Europe is Carlo Ginzburg's The Night Battles: https://en.wikipedia.org/wiki/The_Night_Battles

Thanks! Though I have to say, credit for the open access is entirely due to librarians of the UC system. I just had to click a button saying I wanted it to be made available. The work was already done via this 2019 agreement with Cambridge: https://osc.universityofcalifornia.edu/uc-publisher-relation...

I agree, that story is fascinating. In fact the shift in the 1950s in how scientists thought about psychedelics is actually what my second book (in progress) is about. The Abramson cited in that study is one of the main people I've been researching: https://en.wikipedia.org/wiki/Harold_Alexander_Abramson

So as a follow on, I know this is a long shot but if anyone reading this has family or professional connections to some of the people involved in that first generation of psychedelic research in the 1950s and early 1960s (yes, they are old now, but at least a few are still around) please get in touch with me. I've been conducting oral history interviews for the past year or two and finding out all kinds of interesting things.

Author here: this is an open-access journal article, mostly about the colonial histories of peyote, psilocybin and ayahuasca. If you aren't able to access it and want to, just get in touch (contact info in profile) and I'd be happy to email you a PDF. Same goes with the primary sources I cite, although most should be available via Hathi Trust's emergency temporary access feature.

Very much agree that the UFO thing is a red herring. But perhaps Occam's Razor actually points to an even simpler explanation, that of it being a reoccurring mythological trope that was repackaged as an historical event.

That said, I would love if it actually were the case that a mysterious woman from, say, the Dutch East Indies — to make up a random, semi-plausible example — really did wash ashore in a Japanese fishing village. Sounds like a sequel to "The Thousand Autumns of Jacob de Zoet" (awesome book by the way).

Totally agree. When reading history, you can and should substitute an invisible "According to the limited sources I consulted, and modulo their biases/mistakes/oversights, it seems to me the best interpretation of the given data that..." before every statement. Virtually no professional historian is going to claim to be 100% certain of any interpretation they make. It's often the case that more you dig into sources, the less certain you get. (Which is why almost every academic history paper's argument boils down to "this [person/event/era] turns out to be more complicated than we thought").

Hayden White has written a lot about this, specifically in his book Metahistory. You should check it out if interested! https://en.wikipedia.org/wiki/Metahistory:_The_Historical_Im...

I also really like Fernand Braudel on events as a kind of epiphenomena of history, the misleading surface disturbances underneath the actual, barely-discernible patterns.

Correct. There were two recorded performances, one in December, 1613, the second in January, 1614. I went ahead and described the second because it's better documented. I got the exact date from a journal article by a garden historian. [0] These are the kind of things that footnotes are helpful for - if I were writing this up as an academic paper, I'd get into the weeds with these details, but unfortunately it just doesn't work when you're writing it as a straight-ahead narrative without footnotes. That's especially true because it was basically just an introductory anecdote, not the focus of the piece.

I agree though, digging deep into historical sources, I think, should make us all uncomfortable. As you said, historians should never claim to have direct access to historical truth. It's all mediated and all potentially corrupted by the bias of observers/recorders. That's just a fact of doing history, and it's why we're not humanists, not scientists. It's also why I find it so endlessly interesting.

[0] https://www.jstor.org/stable/25472393?seq=2#metadata_info_ta...

I much prefer footnotes myself, where you can see directly what is being cited. Unfortunately there's a wide range of practices for online writing, but the norm I've found is usually just hyperlinks to digitized sources.

The Appendix, the online history journal I helped create back in the day, tried to experiment with a more detailed way of doing in-line citations, via small icons that you can click to see images, citations of primary source texts, even music or films on occasion. You can see it in action here:

http://theappendix.net/issues/2013/10/made-in-taiwan-an-eigh...

Author here - was it really glorious and strange and beautiful? All subjective judgements, so who can truly say. But I can tell you that those judgements are at least drawn from an actual primary source, published in 1614: the stage directions for the performance itself. The link: https://www.google.com/books/edition/The_Maske_of_Flowers/qC...

Incidentally, that source is linked in the article more than once, along with others.