Linguistic Miscellany Thread

[Vijay]

Do other languages commonly have “false conditionals” of the form “If you’re interested, I’m free next week” (where the speaker is free without regard to the addressee’s interest)?

Ich bin da, wenn du willst?

[akam chinjir]

Mandarin allows it.

Agreeing I think with Sal, I'd say conditionals provide a good way to indicate the relevance of some state of affairs without asserting that the state of affairs obtains. I wouldn't be surprised if lots of languages do it about the same way.

Topicalisation could also work, if there's a way to topicalise a clause and mark it unasserted without just producing a conditional.

Or a question. "Want some beer? There's some in the fridge."

Apparently a survey found that only 50% of people believe modus tollens is valid in indicative conditionals...

Any chance this is an issue of interpretation rather than of bad logic? Ordinary language "if" often gets interpreted as logical "only if," and if the people surveyed were doing that, then what they were rejecting wasn't modus tollens but denying the antecedent (and they were right to do so, of course).

(An example of this sort of interpretation: if you say, "If you want some beer, there's some in the frige," some wag might say, "But what if I don't want any? In that case there's no beer?" ---Of course they're (intentionally) missing your point, but they're not simply misunderstanding the meaning of "if.")

[Salmoneus]

I gather the survey (which I'm just taking from wikipedia, so don't take it as gospel) gave conditionals and asked for what could be known.

So they were given something of the form "if A is true, B is true". Then they were given two situations:
- A is true. Is B true? (everyone says yes)
- B is NOT true. Is A true? (only 50% say no, the other 50% say there's no way to know).

Whereas classical modus ponens insists that in the second case we can know for certain that A is not true (because if it were, then B would be true, and it isn't).

[TurkeySloth]

I have used both *C and <C>, where C represents any character, for orthography. While I know <C> is for orthography, what's the proper use of *C?

[zompist]

So they were given something of the form "if A is true, B is true". Then they were given two situations:
- A is true. Is B true? (everyone says yes)
- B is NOT true. Is A true? (only 50% say no, the other 50% say there's no way to know).

Whereas classical modus ponens insists that in the second case we can know for certain that A is not true (because if it were, then B would be true, and it isn't).

Whenever I hear about such tests, I suspect that people are using subtler tools than the logicians are.

E.g., someone tells you, "If the wire is connected, the light will be on. The light isn't on, so the wire must not be connected." Is that good reasoning?

If you're assigning a logic problem, then yes. In everyday life, it doesn't prove anything at all! The cause may be quite different: the light bulb is out, the power is off, the outlet is broken, the wire you're looking at is the wrong one.

In short, you were told to trust A → B, but in everyday life, the A → B's we hear are usually the weak point of the argument! It's quite right to be skeptical about them.

[gestaltist]

I have used both *C and <C>, where C represents any character, for orthography. While I know <C> is for orthography, what's the proper use of *C?

The asterisk is used to describe an earlier, historical form, or (sometimes) a theoretically possible but incorrect form.

[Raphael]

Silly question: How did "me" in place of "I" become a common feature of stereotypical "caveman speech" in English? Using an inflected version of "I" strikes me as a more complex and sophisticated way to use language than simply using the infinitive!

[Pabappa]

Arguably "me" is the basic form ... c.f. "it's me!" and children saying "me and Daddy went to the store".

[Salmoneus]

Yes, I think "me" is effectively the default, and "I" is the marked form used only in one specific circumstance.

[Salmoneus]

So they were given something of the form "if A is true, B is true". Then they were given two situations:
- A is true. Is B true? (everyone says yes)
- B is NOT true. Is A true? (only 50% say no, the other 50% say there's no way to know).

Whereas classical modus ponens insists that in the second case we can know for certain that A is not true (because if it were, then B would be true, and it isn't).

Whenever I hear about such tests, I suspect that people are using subtler tools than the logicians are.

E.g., someone tells you, "If the wire is connected, the light will be on. The light isn't on, so the wire must not be connected." Is that good reasoning?

If you're assigning a logic problem, then yes. In everyday life, it doesn't prove anything at all! The cause may be quite different: the light bulb is out, the power is off, the outlet is broken, the wire you're looking at is the wrong one.

In short, you were told to trust A → B, but in everyday life, the A → B's we hear are usually the weak point of the argument! It's quite right to be skeptical about them.

True - although that's essentially not understanding the terms of the test. But in any case, that's not all that's happening here, because exactly the same complaint applies to modus ponens. If people are just being skeptical of the truth of "if the wire is connected, the light will be on", they ought to be equally wary of concluding that the light will be on from the mere evidence of the wire being connected. But they're not - they're apparently quick to leap to that conclusion. They're treating the conditional as secure for modus ponens but insecure for modus tollens.

But logically speaking, modus tollens and modus ponens are exactly equivalent in classical logic, because one entails the other. If someone accepts one but not the other, this can't be because of a particular belief (like 'the conditional may not be entirely true'), because that would affect both processes equally. It can only be because they are no thinking in accordance with classical logic (either they're being illogical, or they're following a different logic that classical logic does not capture).

[Ryan of Tinellb]

Dumb question that I'm sure I know the answer to, but it's just not coming to me. What is this construction called? Something like
"his majesty, the king" where two nouns sharing a referent are juxtaposed.
Additional question: are these constructions universal/widespread, or is it a peculiarity of English?

[Travis B.]

Dumb question that I'm sure I know the answer to, but it's just not coming to me. What is this construction called? Something like
"his majesty, the king" where two nouns sharing a referent are juxtaposed.

Apposition.

[Linguoboy]

Dumb question that I'm sure I know the answer to, but it's just not coming to me. What is this construction called? Something like "his majesty, the king" where two nouns sharing a referent are juxtaposed.
Additional question: are these constructions universal/widespread, or is it a peculiarity of English?

I think they're found thoroughout SAE. In Korean they work a little differently, I believe, but I need to check my grammars to make sure.

[Raphael]

Another silly question: what's infinite about infinitives?

[Moose-tache]

"Finite" verbs are verbs limited to a specific referent, like Latin "amant" or English "loves," while verb forms that are not limited in this way are in-finite, like "amare" or "to love." It makes less sense when you apply it to other languages (Does Japanese "wakarimashita" seem any more finite than "wakaru?"), which is why sometimes the term lemma will be used, or even something awkward like "dictionary form" for languages that don't have IE-style conjugations. The term is further complicated by the fact that in some languages, in IE, Semitic, and other families, the infinitive is nearly indistinguishable in form and usage from a gerund.

[Raphael]

"Finite" verbs are verbs limited to a specific referent, like Latin "amant" or English "loves," while verb forms that are not limited in this way are in-finite, like "amare" or "to love." It makes less sense when you apply it to other languages (Does Japanese "wakarimashita" seem any more finite than "wakaru?"), which is why sometimes the term lemma will be used, or even something awkward like "dictionary form" for languages that don't have IE-style conjugations. The term is further complicated by the fact that in some languages, in IE, Semitic, and other families, the infinitive is nearly indistinguishable in form and usage from a gerund.

Thank you!

[TurkeySloth]

I have used both *C and <C>, where C represents any character, for orthography. While I know <C> is for orthography, what's the proper use of *C?

The asterisk is used to describe an earlier, historical form, or (sometimes) a theoretically possible but incorrect form.

Okay. Thanks.

[Richard W]

I have used both *C and <C>, where C represents any character, for orthography. While I know <C> is for orthography, what's the proper use of *C?

The asterisk is used to describe an earlier, historical form, or (sometimes) a theoretically possible but incorrect form.

Okay. Thanks.

The common idea is 'unattested, i.e. no examples', but that gets confusing with attested forms universally agreed to be blunders. Double asterisk is sometimes used for 'grammatically incorrect', i.e. something no-one exercising native competence would utter.

[magb]

I got garden pathed hard by a sentence in this Reddit post about Saudi prisons for women:

Women in these prisons are often victims of abuse, rape, attempted runaways, activists, women who try to divorce their husbands, and women who are seen as generally rebellious by their parents or male guardian.

My journey through the garden went like this:

"Women in these prisons are often victims of abuse"
Yes, I'm sure there's a lot of abuse going on in those prisons.

"...are often victims of abuse, rape"
Sadly yes.

"...are often victims of abuse, rape, attempted runaways"
It's a little odd to describe them as being "victims" of trying to run away, but I get what they're trying to say.

"...are often victims of abuse, rape, attempted runaways, activists"
Wait, what? Victims of activists?

"...are often victims of abuse, rape, attempted runaways, activists, women who try to divorce their husbands"
Ohhhhhh.

Of course, what messed me up here was the comma between "abuse" and "rape". The sentence would've been less confusing if it instead read "...victims of abuse, victims of rape, ..." or "...victims of abuse and/or rape, ..." But I still found it weird how long it took me to figure out the intended syntax.

[Xwtek]

What linguistic abbreviation to be used if my language has a suffix that means either 1SG or 2SG. (i.e. German except German merges 1SG and 3SG instead of 1SG and 2SG)