Linguistic Miscellany Thread

[Richard W]

Well, they're a special type of construction, where both the subject and subject complement are clauses, and their order is reversed.

To us English speakers, they're the natural order, and we wouldn't reverse the identification. Indeed, as the first phrase is more specific, I would say that it was the subject. They also follow the principle of having lengthy parts at the end of the sentence.

[jal]

To us English speakers, they're the natural order, and we wouldn't reverse the identification. Indeed, as the first phrase is more specific, I would say that it was the subject. They also follow the principle of having lengthy parts at the end of the sentence.

I know how it works in English (and maybe you're right with regards to the syntactic subject), but that wasn't my question. This isn't the English thread. I was solliciting knowledge about languages other than English.


JAL

[malloc]

Does anyone else feel like so many letters of the Hebrew alphabet look remarkably similar to the point of confusion? Consider pairs like bet and kaf, dalet and resh, or he and tav, to say nothing of the word-final variants. Along with the well-known absence of vowels in Hebrew writing, this seems like it would make reading Hebrew (or other languages written in this alphabet) rather challenging.

[Richard W]

Does anyone else feel like so many letters of the Hebrew alphabet look remarkably similar to the point of confusion? Consider pairs like bet and kaf, dalet and resh, or he and tav, to say nothing of the word-final variants. Along with the well-known absence of vowels in Hebrew writing, this seems like it would make reading Hebrew (or other languages written in this alphabet) rather challenging.

It's a well-known problem of the main family of Semitic alphabets. Later alphabets permanently merged phonetically distinct letters. I think the worst offender is the writing system for Avestan. The standard Mongolian writing system, which has become alphabetic, does distinguish the letters, but one has to keep track of the context.

Aramaic distinguishes I think 3 consonants in speech but not writing - at some point the 3 pairs changed, so there was clearly an unwritten distinction.

Arabic suffered from a loss of distinction, as seen in the main Kufic style. The dots that nowadays distinguish many letters are a later addition, not to be confused with the vowel marks.

[bradrn]

Does anyone else feel like so many letters of the Hebrew alphabet look remarkably similar to the point of confusion? Consider pairs like bet and kaf, dalet and resh, or he and tav, to say nothing of the word-final variants. Along with the well-known absence of vowels in Hebrew writing, this seems like it would make reading Hebrew (or other languages written in this alphabet) rather challenging.

It does get confusing sometimes, but from the perspective of someone who needs to read Hebrew fairly regularly, I’d say that it doesn’t seem to be a huge problem in practice. It depends on typeface too: for instance, that used by Sefaria is probably less liable to misreading than that used by Chabad, although both are still very easy to read for me. There’s a more traditional typeface which is worse than both, but I can’t seem to find any images of it… [EDIT: here]

Another thing to note is that handwritten Hebrew (of the sort erroneously referred to as ‘cursive’) is much less ambiguous. Of the letter pairs you mention, bet gets an extra curve, dalet gets a loop, and he is (at least for some people) altered to two concentric arcs.

(Also, I’m surprised you mention he/tav, and not he/chet — the latter pair are far more similar to each other than the former.)

Does anyone else feel like so many letters of the Hebrew alphabet look remarkably similar to the point of confusion? Consider pairs like bet and kaf, dalet and resh, or he and tav, to say nothing of the word-final variants. Along with the well-known absence of vowels in Hebrew writing, this seems like it would make reading Hebrew (or other languages written in this alphabet) rather challenging.

It's a well-known problem of the main family of Semitic alphabets. Later alphabets permanently merged phonetically distinct letters. I think the worst offender is the writing system for Avestan. The standard Mongolian writing system, which has become alphabetic, does distinguish the letters, but one has to keep track of the context.

I’d say the worst offender is not Avestan, but its parent Pahlavi, which famously merged 24 phonemes into 13 letters, essentially becoming a logography.

[malloc]

I’d say the worst offender is not Avestan, but its parent Pahlavi, which famously merged 24 phonemes into 13 letters, essentially becoming a logography.

Now that's mind-boggling.

Zompist once advised keeping letters in your conscript maximally distinct and observed that the elves in LotR must have horrible levels of dyslexia given how similar tengwar letters look. I often get the same impression from Hebrew and Arabic.

[Raphael]

I’d say the worst offender is not Avestan, but its parent Pahlavi, which famously merged 24 phonemes into 13 letters, essentially becoming a logography.

Now that's mind-boggling.

Zompist once advised keeping letters in your conscript maximally distinct and observed that the elves in LotR must have horrible levels of dyslexia given how similar tengwar letters look. I often get the same impression from Hebrew and Arabic.

OK, but how does the Latin alphabet look like to people who are used to Arabic or Hebrew?

[Lērisama]

I’d say the worst offender is not Avestan, but its parent Pahlavi, which famously merged 24 phonemes into 13 letters, essentially becoming a logography.

Now that's mind-boggling.

Zompist once advised keeping letters in your conscript maximally distinct and observed that the elves in LotR must have horrible levels of dyslexia given how similar tengwar letters look. I often get the same impression from Hebrew and Arabic.

I think it's more about unfamiliarity than similarity, but you learned the Latin script too long ago to remember that feeling with it. I don't think there's any script where all the letters look distinct to someone illiterate in it.

Edit: what Raphael said

[Raphael]

I don't think there's any script where all the letters look distinct to someone illiterate in it.

Perhaps not in real life, but that might make for an interesting conscripting challenge...

[zompist]

I don't think there's any script where all the letters look distinct to someone illiterate in it.

I don't know about that— the Greek alphabet has pretty distinct letterforms. So far as I can see it doesn't even have mirrored letterforms, cf. b/d and p/q. Nor can any majuscules be confused with minuscules, cf. capital I / little l.

[bradrn]

I don't think there's any script where all the letters look distinct to someone illiterate in it.

I don't know about that— the Greek alphabet has pretty distinct letterforms. So far as I can see it doesn't even have mirrored letterforms, cf. b/d and p/q. Nor can any majuscules be confused with minuscules, cf. capital I / little l.

What about nu/ypsilon, zeta/xi, or in some styles phi/psi?

EDIT: also sigma/omicron

[Civil War Bugle]

I took several courses in Arabic in college (longer ago than I want it to be) and took one semester of Hebrew. I was one of two non-Jewish people in the Hebrew class, and I think the professor designed the syllabus on the assumption that everyone in the class had learned the alphabet at some point before attending college, e.g. while preparing for bar mitzvahs or what have you. So the course sped through that aspect, and I spent a week or so trying to catch up on the reading aspect, and doing a certain amount of guesswork based on cognates, but once I did catch up, it seemed fine from the perspective of a literate person.

On the other hand, having never applied what I learned in that class since graduation, my memory of the Hebrew alphabet has atrophied a bit (not entirely but...) while I still can read and produce the entirety of the Arabic character set as used in Arabic.

[zompist]

I don't know about that— the Greek alphabet has pretty distinct letterforms. So far as I can see it doesn't even have mirrored letterforms, cf. b/d and p/q. Nor can any majuscules be confused with minuscules, cf. capital I / little l.

What about nu/ypsilon, zeta/xi, or in some styles phi/psi?
EDIT: also sigma/omicron

I don't see those as very similar. It's not hard to distinguish pointy from round (ν/υ) or open from closed (ψ/φ). I'll grant that ο/σ are close enough to require some learning. I'd also point out that these similarities apply only to the minuscules— as we've seen, long centuries often make letterforms approach each other.

I find ח/ת or ע/ﬠ or דר or כ/ב much more confusing. On the other hand, modern fonts help a lot. E.g. when כ is circular it's easily recognized.

[bradrn]

I find ח/ת or ע/ﬠ or דר or כ/ב much more confusing. On the other hand, modern fonts help a lot. E.g. when כ is circular it's easily recognized.

Firstly, ע/ﬠ are not even separate letters in the first place. In fact I didn’t even know that that ‘alternative ayin’ existed at all — looking at the Unicode standard, it seems to be intended for use with diacritics below.

As for the rest, I’ve never had a problem with ת/ח, which look pretty different to me. Far more confusing is ח/ה, which in some traditional typefaces can look nearly identical. ר/ד and כ/ב can be somewhat confusing, but I feel that typeface design largely manages to keep them separate.

Another one you didn’t mention is ס/ם. Most modern typefaces are good at keeping them separate, but traditionally they could look similar enough that I have gotten confused, though it’s ameliorated by the fact that ם is restricted in distribution.

[zompist]

I find ח/ת or ע/ﬠ or דר or כ/ב much more confusing. On the other hand, modern fonts help a lot. E.g. when כ is circular it's easily recognized.

Firstly, ע/ﬠ are not even separate letters in the first place. In fact I didn’t even know that that ‘alternative ayin’ existed at all — looking at the Unicode standard, it seems to be intended for use with diacritics below.

Got the wrong letters-- I meant ע/צ.

As for the rest, I’ve never had a problem with ת/ח, which look pretty different to me. Far more confusing is ח/ה, which in some traditional typefaces can look nearly identical. ר/ד and כ/ב can be somewhat confusing, but I feel that typeface design largely manages to keep them separate.

Another one you didn’t mention is ס/ם. Most modern typefaces are good at keeping them separate, but traditionally they could look similar enough that I have gotten confused, though it’s ameliorated by the fact that ם is restricted in distribution.

A modern font can help a lot! I blame calligraphers... they seem to like letters to look alike. Blackletter (for the Roman alphabet) is just as bad.

[bradrn]

A modern font can help a lot! I blame calligraphers... they seem to like letters to look alike. Blackletter (for the Roman alphabet) is just as bad.

The difficulty is that homogeneous scripts tend to be easier to write. As annoying as blackletter is to read, it’s marvellously efficient in both space and time when written with the appropriate pen.

(I’ll also note that ‘blackletter’ is a rather broad term. Textura is significantly more homogeneous than Fraktur.)

[Man in Space]

I find the more traditional type is easier to read than modern sans-serif fonts or handwriting—the newer stuff omits the things that help me distinguish certain letters by stroke (and in some cases, most notably aleph, the newer style looks completely unintuitive—I tend to write aleph using three strokes, to wit one small right-to-left diagonal, a large left-to-right diagonal, and then a third right-to-left, not an arc and vertical line).

Arabic (well at least naskhi) gives me similar problems.

[bradrn]

I find the more traditional type is easier to read than modern sans-serif fonts or handwriting—the newer stuff omits the things that help me distinguish certain letters by stroke (and in some cases, most notably aleph, the newer style looks completely unintuitive—I tend to write aleph using three strokes, to wit one small right-to-left diagonal, a large left-to-right diagonal, and then a third right-to-left, not an arc and vertical line).

The thing is, printed Hebrew and modern Hebrew are essentially two different scripts now. I remember learning them separately in primary school as such.

[Richard W]

Our family's been using human pronouns and classifier(s) when talking about our kitten in Thai. One such pronoun is หนู, widely used as 2nd/3rd person pronoun for youngsters. Yesterday it finally actually happened - my wife said หนูเป็นแมว. While she meant, "She's a cat", it can also translate as "The mouse is a cat"!

[alice]

According to something I heard on the radio yesterday, "calque" is a loanword, whereas "loanword" is a calque. How many other languages have this, or similar?