CONLOQVIVM

Sebastian Koppehel

01:55

Ah, the joy of the "Hot Network Questions."

Not only got this utterly banal answer 29 upvotes, but I was also for the first time asked by a coworker if I happened to know Latin -_-

Cerberus

@SebastianKoppehel It's kind of interesting and unusual, so not exactly banal!

What did you say to your collega?

Sebastian Koppehel

02:15

@Cerberus I know a bit ;)

Cerberus

@SebastianKoppehel Such a braggard!

Cerberus

02:50

> The first instance of synizesis in the Aeneid is in Book 1, line 8, where “Italiam” is pronounced as two syllables instead of three. The line reads as follows:

Lā.vī.ni.aque vē.nit lī.tor.a mult.um il.le et terr.is iac.tā.tus et alt.o

(Lavini.aque venit lit.or.a mult.um ille et terr.is iactat.us et alt.o)

The word “Italiam” is contracted into “itali.am” and forms one syllable with the following word “que”. This makes the line fit into the dactylic hexameter pattern. Synizesis here helps create a smooth transition between the words and avoids a hiatus.

Look, how Chat GPT is progressing.

However, this was in the middle of a conversation; when I ask in a new conversation, it actually produced the correct answer.

But then it failed again later.

Producing a string of weird mistakes.

cmw

04:03

@SebastianKoppehel I've complained of it before, but it's a fact of life, and honestly, after a certain point, the aggregate votes don't really matter.

Interesting that they recognized you from it, though.

Canned Man

13:46

This may be a surprisingly simple question, but nevertheless here goes:
In LLPSI cap. VIII ll. 137ff, the following is written: ‘Quis saccum portat? Servus saccum portat. Quī servus? Servus quī saccum portat est Syrus. Is/ille servus saccum portat.’
I don’t understand why the second question uses ‘quī’ rather than ‘quis’. Is this Ørberg trying to show that the two in practice were interchangable? The same applies to ll. 184f with quid/quod.

cmw

14:44

@CannedMan It's the interrogative adjective here, rather than the interrogative pronoun.

"Who carries the sack? The slave carries the sack. Which slave? ...Syrus."

Canned Man

Right, these are in fact not differentiated in my grammar; they are both listed under interrogative pronouns.

There is a comment mentioning that quod is always used as adjectivically (is that even a word?), but no a clear differentiation as you show.

cmw

Adjectivally, I think. Interesting! What grammar are you using? Is it Norwegian?

Basically, when quis, quid modifies a noun, it takes the form of qui, quae, quod.

Canned Man

15:03

I have two: a short grammar for high school (Norwegian) and a detailed grammar for university (Swedish). (I haven’t checked the Swedish entry yet.)

In the _Exercitia_, for the phrase ‘Qu__ Rōmam it? Servus Rōmam it. Qu__ servus? Servus qu__ Rōmam it est Mēdus, servus improbus. Is servus Rōmam it.’, the solution is listed as quis, quī/quis and quī. So: ‘Quis Rōmam it? … Quī/quis servus? Servus quī Rōmam it est Mēdus.’

cmw

Yep, right.

The first is unquestionably an interrogative pronoun. The second is an interrogative adjective, and the last is a relative pronoun.

Canned Man

Why then does he allow both options in the answer key?

The grammars I use are:
Samson Eitrem: _Latinsk grammatikk,_ third edition, Aschehoug 1996.
Nils Sjöstrand: _Ny latinsk grammatik,_ Gleerups förlag, Lund / Nya litografen, Malmö 1960.

Sjöstrand does have a separate section for interrogative adjectives, but there only lists quālis, quantus, quantum and quot. Quis/quī, quae, quid/quod are listed under interrogative pronouns. Are we here faced with different grammatical traditions?

cmw

15:54

Even the ancients weren't consistent.

@CannedMan Here it is explained in Allen and Greenough: dcc.dickinson.edu/grammar/latin/…

I think it's useful to make the distinction, even if it isn't consistently applied.

Adam

18:07

@CannedMan I remember that line!

I keep meaning to post a question about different passage in LLPSI.

tl;dr, It's the usage of pronouns like is in this passage:

> Dies mensis primus 'kalendae' nominatur. Dies primus mensis Ianuarii dicitu 'kalendae Ianuariae'; is dies anni primus est atque initium anni novi.

Oh, it's my English brain wanting to put importance on word order.

It's "he is the first day of a year and also the start of a new year"

I wanted to apply English word order and say something like, "he, the day, is the first of the year", and the pronoun felt superfluous.

Cerberus

@cmw I would say quis is uncommon adjectivally, nor is it super rare.

But quid adjective? I don't remember ever seeing that.

> GPT: “Fortius” is the neuter form of the comparative adjective “fortior”, which means “stronger”. In this context, it is used to describe the bull. The neuter form is used because the gender of the noun “taurus” (bull) is masculine, and the adjective must agree in gender with the noun it modifies.

Yup makes sense.

cmw

18:26

@Cerberus It actually does! It's really bad at performing calculations apparently, and so it can't "logically intuit" what it says. It's just parroting, and it messes up because it's putting two sentences together that don't belong.

I tried having it evaluate some numerical pattern sequences, and it failed hard.

(Except for Fibonacci's sequence.)

Cerberus

@cmw Yes, in that way, it makes sense.

@cmw Oh that's interesting, even something as simple as that.

But they are linking to Wolfram Alpha.

That should make it much better as simple mathematics and logic.

GPT is good at translating language into logical instructions, and then Wolfram does the rest.

cmw

@Cerberus Yeah, once they do that, all hell will break loose. ;P

Cerberus

@cmw Yay!

I'll try to keep things under control, but I can't guard all of it at once.

cmw

Eat as many souls as you can.

Cerberus

I will try.

Do AIs have souls?

cmw

19:01

Not yet.

Ever see Talos in Hades?

Joonas Ilmavirta

@cmw The OEIS is a nice tool if you really need to identify an odd sequence.

@Cerberus Have you never had students write such things in exams? They can get quite incoherent when lost and pressed for time.

@Adam That example shows well that it's not good to see is as "he". The ancients did personify things, but that's not what's going on here.

Adam

19:18

@JoonasIlmavirta True; it's just matching the gender of the noun it represents, which in an English translation should use it rather than he.

cmw

19:29

@JoonasIlmavirta It uses a database, it seems.

Joonas Ilmavirta

@cmw Yes, it does. But it is an extensive one. If you plug in just a few numbers, it gives hordes of options. It requires some work to use for good, but it can be a source of inspiration for stuff like that.

@Adam Or you could just use an emphatic "the" or "the/this very", a device Latin doesn't have for emphasis.

Incidentally, Finnish doesn't have gender, but we do differentiate "he/she" from "it". However, in spoken language it's very common to refer to people as "it", to the extent that it really sticks out if you use an actual personal pronoun for a person.

@Adam I think so. My impression is that there will be a number of mostly independent teams working on different projects, but I could be off. They will have some degree of internal coordination no matter what.

cmw

"Sorry, but the terms do not match anything in the table."

Oh, wait, I made a mistake!

Let me double check before I gloat.

Joonas Ilmavirta

@cmw Surprising! Did you input integers only?

cmw

Nevermind! It got me!

I typoed.

Tried another, it got it. It's good!

Adam

@JoonasIlmavirta Interesting! Does that mean that there were no language evolutions to support different gender identities? I recall reading that the single they/them in English is a somewhat recent evolution within my lifetime.

cmw

19:43

@Adam That would be incorrect. Shakespeare has a singular they. itre.cis.upenn.edu/~myl/languagelog/archives/002748.html

Joonas Ilmavirta

@Adam It seems that this language family never had gender to begin with.

cmw

I think Geoffrey Pullum made a mistake here, though. The gender isn't known! The indefinite "man" could still refer to a woman.

Joonas Ilmavirta

English doesn't really have much gender. It's only in those pronouns and the meanings of words (like "fireman"), not all nouns having gender like in some other languages.

cmw

English used to, though. But Finnish never did, nor does any of the Finno-Uralic languages.

Joonas Ilmavirta

@cmw Yup, it's not something we forgot but something we never had.

Adam

19:48

@cmw Oh ho ho!

Joonas Ilmavirta

It's pretty clear that a predecessor of modern English must have had gendered nouns, but I never looked into it and don't know whether it happened in anything we might call English.

cmw

@JoonasIlmavirta Modern English has traces, but no, you have you go to Old English for it.

Cerberus

20:03

@JoonasIlmavirta Not this bad, I think!

@Adam It's even worse than that, and I quote:

> The neuter form is used because the gender of the noun “taurus” (bull) is masculine

Adam

It's really interesting how much closer Old English is to other languages descended from PIE.

Joonas Ilmavirta

20:20

@Cerberus Interesting! I find it somehow fascinating to understand the unique ways in which people and computers err. They appear quite different.

Cerberus

@JoonasIlmavirta Indeed!

Human errors seem more praedictable to us.

Joonas Ilmavirta

@Adam Not a surprise. When things diverge from a common origin, you expect similarity to increase when looking further back. In fact, that time dependence of similarity is a strong clue of a common origin.

@Cerberus I guess that when humans err, their thinking makes some vague sense and we can follow it, but we can't follow an almost-sensical computer. The metrics of what makes an argument almost valid are different, even if both agree on what a fully sound argument looks like.

Cerberus

21:25

@JoonasIlmavirta Yes. And I would say the computer is nowhere near sensical?

Joonas Ilmavirta

22:30

@Cerberus I would agree, but we're using a very human metric on proximity to sense. The computer may care more about a thing having the general feeling of a syllogism than having the parts actually logically connect, so I am very tempted to say that the computer thinks it's close to making sense.

Cerberus

@JoonasIlmavirta Sure, I meant from a human perspective, from the perspective of, how useful is this reply for me?

Transcript for