Using ChatGPT

[Peter Kirby]

Given that "... posts here should conform to the norms of academic discussion....,"
is use of ChatGPT (cited after the fact) here appropriate?

I moved the ChatGPT generated stuff and the following to this thread.

Apart from limited use where the author isn't just quoting some output but has integrated it into their own thought process and outline (e.g., using a translation), I have been trying to move all ChatGPT-generated output here (even when posted to Christian Texts & History).

[billd89]

Well, the opinions of the peanut gallery here are of no more value than my own. I would rather an objective arbiter of the truth: perhaps AI will provide that? If not today (2024), soon.

I asked ChatGPT 3.5 if it could date a Latin passage to determine whether said passage came from the 1st, 2nd or 3rd C AD. The program insists it can! While I don't automatically believe I also don't necessarily disbelieve: let's see what it says about the Latin fragment above.

This passage appears to be a historical account discussing the situation in Egypt, particularly mentioning the presence of Christians and their practices in relation to the worship of Serapis (also spelled Serapidem). The use of "Chrestus" or "Chrestianus" instead of "Christus" or "Christianus" is often considered a characteristic feature of early Christian writings, as it reflects the phonetic similarity between the Greek word for "good" or "useful" (χρηστός, chrestos) and the name of Christ.

Based on the content and language used in the passage, it suggests a date likely within the 2nd or 3rd centuries AD. During this time, Christianity was spreading and developing, and there was still a fluidity in terminology and religious practices. The absence of certain titles and the interchangeability of terms like "Chrestus" and "Christus" are indicative of an early period when Christianity was still forming as a distinct religious identity.

Therefore, this passage is likely from the Second or Third Centuries AD.

Within a few years, I suppose we will have some definitive and very reliable answers that supersede (and blow away) whatever "conclusions" some tired egos -- legends in their own minds -- here would imagine (and insist) settled.

Tant Pis.

I'm uncertain. I'm also a Doubting Thomas.

Me:

Could the letter be a later forgery? Are there any clues the author was actually an Imperial authority?

ChatGPT
While it's theoretically possible that the letter could be a later forgery, there are several reasons to believe it is authentic:

1. Style and Content Consistency: The style and content of the letter are consistent with other known writings of Hadrian and typical Roman imperial correspondence. The themes discussed, such as the situation in Egypt and Hadrian's observations on society and religion, align with what is known about Hadrian's interests and concerns.

2. Historical Context: The events and references mentioned in the letter fit within the historical context of Hadrian's reign. Hadrian was known for his extensive travels throughout the empire, including visits to Egypt, and he took a keen interest in the affairs of the provinces.

3. Language and Tone: The language and tone of the letter reflect the formal and diplomatic style typical of Roman imperial correspondence. Additionally, the use of specific titles and references, such as "Augustus" and "consul," further suggest the authority of the author as an imperial figure.

4. Manuscript Evidence: If there are existing manuscripts or copies of the letter that date back to antiquity and are consistent with the historical context, language, and style of Hadrian's era, it would provide further evidence of its authenticity.

However, without specific historical documentation or additional evidence confirming the authenticity of this particular letter, absolute certainty cannot be guaranteed. Scholars often analyze various factors, including linguistic analysis, historical context, and manuscript evidence, to determine the authenticity of ancient texts.

[Peter Kirby]

To be more clear, any post that relies on ChatGPT to provide conclusions (even when stated provisionally) will be moved to this thread.

Posters may use ChatGPT to find ideas if they need to, but make them their own.

[billd89]

Oh, I'm very clear this is what ChatGPT says in reply to my own specific questions.

Me:

Could ChatGPT 3.5 estimate the probability (50%? 80%? any percentage) that last passage was written by an Imperial authority of the 2nd Century?

ChatGPT
Based on the information provided and considering the style, content, and historical context of the passage, it's highly likely that the last passage was written by an Imperial authority of the 2nd century, particularly Emperor Hadrian. Given the consistency with known writings of Hadrian and the thematic alignment with his interests and concerns, I would estimate the probability to be around 90-95% that the passage was written by an Imperial authority of the 2nd century. However, without specific manuscript evidence or further historical analysis, it's challenging to provide an exact percentage.

[billd89]

They aren't entirely useless. To store up something necessarily means to store up something gor its later use. The list represents a paraphrase of each citation.

"The list represents a paraphrase of each citation." Does it though? I'm unclear if this is ACTUALLY a 'Garbage-In/Garbage-Out' situation or smthg that rationale minds would agree 'debatable.' I think "hallucinations" is outright garbage, if it's simply Error. I don't mean one little bit, a few dubious points here and there -- but rather pages of examples. In other words: ALL that sh*t needs to be double-checked.

Given that "... posts here should conform to the norms of academic discussion....,"
is use of ChatGPT (cited after the fact) here appropriate?

Lots and lots to unpack there, Stephen. But the simple answer should be: NO, AI-generated speculations need to be labeled carefully.

El Pais is a respectable news source. I saw this disturbing image yesterday; wtf?! Does anyone think wheeled-carriage vehicles that scale -- like that -- actually existed in ancient Rome? Definitely an hallucination!

[Peter Kirby]

Oh, I'm very clear this is what ChatGPT says in reply to my own specific questions.

Me:

Could ChatGPT 3.5 estimate the probability (50%? 80%? any percentage) that last passage was written by an Imperial authority of the 2nd Century?

ChatGPT
Based on the information provided and considering the style, content, and historical context of the passage, it's highly likely that the last passage was written by an Imperial authority of the 2nd century, particularly Emperor Hadrian. Given the consistency with known writings of Hadrian and the thematic alignment with his interests and concerns, I would estimate the probability to be around 90-95% that the passage was written by an Imperial authority of the 2nd century. However, without specific manuscript evidence or further historical analysis, it's challenging to provide an exact percentage.

Right. I guess I was not clear.

Quoting ChatGPT will land a post in this thread.

Exceptions will be considered on a case by case basis. One known exception is using ChatGPT to do a translation.

(This applies to all LLMs.)

[billd89]

Thematic examples which are "useful" for this thesis:

...

Philo of Alexandria, "De Vita Contemplativa" (1.478):
"...τὰ μὲν ἄχρηστα καὶ τὰ μηδὲν ἀποκειμένα πρὸς χρῆσιν..."
Translation: "...the useless and those laid up for no use..."
Context: Describes things as either useless or laid up for no practical use.

They aren't entirely useless. ...The list represents a paraphrase of each citation.

But where is this citation "De Vita Contemplativa" (1.478) ??? Not only is the quoted fragment bogus (those words don't even appear in DVC!), the citation is made-up too lol

It's utterly bonkers, tho. It's like your computer decided to generate a ream of lies for you, because it knows you're so dim and/or lazy you won't even bother to check.

[DCHindley]

Is everyone blind? "The useless and those laid up for no use" apparently has DVC 1.47-48 in mind:

47 and in this manner they exist to the very end of their lives, without a house and without a home, the enemies of their parents, and of their wives, and of their children, and the enemies of their country, and the worst enemies of all to themselves. For a debauched and profligate life is apt to lay snares for every one.

48 And perhaps some people may be inclined to approve of the arrangement of such entertainments which at present prevails everywhere, from an admiration of, and a desire of
imitating, the luxury and extravagance of the Italians which both Greeks and barbarians emulate, making all their preparations with a view to show rather than to real enjoyment,

In my opinion, section 47 (in red) relays an observation of Philo himself from an old notebook he had with him as a laddy visiting the Mariotes lake resort area as a tourist - where he saw and admired these crazy hipsters. This seems to be a misplaced comment of some kind I may have accidentally attributed to Philo (as an editor of his own notebooks). It has the tone of the worldly editor, not the youthful positivism of young Philo the verbal sketch artist.

Section 48 is a comment by a person I identify as a worldly final editor/publisher bold enough to add his own take on the subject being relayed by Philo. If that is also what sect 47 is, a comment by the final editor, that would make sense. I guess I, like Giuseppe, can make mistakes and admit them too.

Sciafe export of DVC 1.47-48:

47 τοῦτον τὸν τρόπον διαζῶντες ἄοικοι καὶ ἀνέστιοι διατελοῦσιν, ἐχθροὶ μὲν
γονέων καὶ γυναικῶν καὶ τέκνων, ἐχθροὶ δὲ καὶ τῆς πατρίδος, πολέμιοι
δὲ καὶ ἑαυτῶν· ὑγρὸς γὰρ καὶ ἄσωτος βίος ἅπασιν ἐπίβουλος.

48 Ἴσως δ’ ἄν τις ἀποδέξαιτο τὴν ἐπιπολάζουσαν νυνὶ τῶν
συμποσίων πανταχοῦ διάθεσιν κατὰ πόθον τῆς Ἰταλικῆς πολυτελείας
καὶ τρυφῆς, ἣν ἐζήλωσαν Ἕλληνές τε καὶ βάρβαροι πρὸς ἐπίδειξιν μᾶλλον
ἢ πρὸς εὐωχίαν ποιούμενοι τὰς παρασκευάς.

Of course this has nothing to do with the text cited in AA's quote of "Chat GPT" output. Maybe that is AA's point, yes?

It looks good, but it is "a view to show rather than to real enjoyment." Golly, isn't that the point of the comment in DVC 1.48? What a coincidence.

DCH

[billd89]

erm, not really. I don't follow you at all, nor any of these 'hallucinations.'
Sorry I don't 'do' Rorschach interpretations: I need abit more certainty.

I'm still rather confused where ChatGPT starts and finishes, so I asked again.

Part 1: I started by feeding the e-Beast that Philonic DVC 1.47-8 passage, for which I received an adequate trans.:

Chat GPT:
"Thus, living in this manner, they continue to be homeless and without a city, enemies of their parents, wives, and children, enemies also of their own country, and adversaries even to themselves; for a moist and dissolute life is treacherous to all.* Perhaps someone might recognize the current trend prevailing everywhere in the banquets, characterized by a desire for luxurious living and indulgence, which both Greeks and barbarians have emulated, seeking to display their preparations more for ostentation than for enjoyment."

*This phrase appeals to me in an altogether different way, but 'usefulness' from the cited verb ἀπόκειμαι (to be laid away, be laid up in store) is in no way implied. Unfortunately, 'Dissolute' (or: hopelessly sick) is not identical to 'Uselessness' either: 'close, but no cigar!'

Part 2: The next task was to figure out Secret Alias' apparently gibberish reference, paraphrase, or whatever: @#$$#%&**(&%&&!!!.

Me:
where is this phrase from? "...τὰ μὲν ἄχρηστα καὶ τὰ μηδὲν ἀποκειμένα πρὸς χρῆσιν..."

ChatGPT:
The phrase "τὰ μὲν ἄχρηστα καὶ τὰ μηδὲν ἀποκειμένα πρὸς χρῆσιν" translates to "the useless and the things that are of no use at all" in English. This phrase is from the works of the ancient Greek philosopher Plato, specifically from his dialogue "Theaetetus." It is part of a passage where Socrates discusses the nature of knowledge and perception with Theaetetus.

To my brief follow-up query, ChatGPT 3.5 specified this citation: "Plato, Theaetetus, 155d–156a."

Going to Perseus (155d), we DO NOT find this text "...τὰ μὲν ἄχρηστα καὶ τὰ μηδὲν ἀποκειμένα πρὸς χρῆσιν..." nor any Socratic rhetoric on Uselessness, here. (The Perseus trans. is identified as from Plato. Plato in Twelve Volumes, Vol. 12 translated by Harold N. Fowler. Cambridge, MA, Harvard University Press; London, William Heinemann Ltd. 1921.)

Socrates:
Theodorus seems to be a pretty good guesser about your nature. For this feeling of wonder shows that you are a philosopher, since wonder is the only beginning of philosophy, and he who said that Iris was the child of Thaumas1 made a good genealogy. But do you begin to understand why these things are so, according to the doctrine we attribute to Protagoras, or do you not as yet?

Theaetetus
Not yet, I think.

Socrates
And will you be grateful to me if I help you (155e) to search out the hidden truth of the thought of a famous man or, I should say, of famous men?

Theaetetus
Of course I shall be grateful, very grateful.

Socrates
Look round and see that none of the uninitiated is listening. The uninitiated are those who think nothing is except what they can grasp firmly with their hands, and who deny the existence of actions and generation and all that is invisible.

Theaetetus
Truly, Socrates, those you speak of are very stubborn [156a] and perverse mortals.

Of course this has nothing to do with the text cited in AA's quote of "Chat GPT" output. Maybe that is AA's point, yes?

SA's Chatbot point? No frikken clue! I'm not sure how he got ChatGPT (whichever version??) to generate that Greek phrasing which I also cannot find online. There's nothing sensible nor coherent to grasp here: his Chatbot output is literally like talking to the most fucked-up alkie at the dive bar about ... oh ... anything serious. If you take it at face-value -- HOO-BOY!

Me:
Is that relevant to the other, longer Greek passage which I provided?

ChatGPT:
Apologies for the confusion. No, the citation I provided is not directly relevant to the longer Greek passage you provided earlier. The longer passage appears to be from a different source, not from Plato's "Theaetetus." If you have any other questions or if you need further assistance, please feel free to ask.

[andrewcriddle]

Oh, I'm very clear this is what ChatGPT says in reply to my own specific questions.

Me:

Could ChatGPT 3.5 estimate the probability (50%? 80%? any percentage) that last passage was written by an Imperial authority of the 2nd Century?

ChatGPT
Based on the information provided and considering the style, content, and historical context of the passage, it's highly likely that the last passage was written by an Imperial authority of the 2nd century, particularly Emperor Hadrian. Given the consistency with known writings of Hadrian and the thematic alignment with his interests and concerns, I would estimate the probability to be around 90-95% that the passage was written by an Imperial authority of the 2nd century. However, without specific manuscript evidence or further historical analysis, it's challenging to provide an exact percentage.

This result does not improve my opinion of ChatGPT.

Andrew Criddle