TIL: Today I Learned

Status
Not open for further replies.
Proto-Indo European had declensions so all IE Languages exhibit declension in some form or another, even English. An IE language's words are, traditionally, composed of three parts: a root word - usually a verb existing in a verbal form - plus a particle suffix indicating how the verb's action is being expressed. These two pieces join to form the stem. The Latin term doctor is an example of a stem - the verb doceo ("I teach, educate, bring up") is joined with the masculine agent particle -tor ("He who") to get the word doctor "one who teaches". To this stem is attached inflectional endings which follow a prescribed pattern according to the shape of the word. Syntax is divided into a number of cases, each tasked with indicating an individual word's role in the wider context of the clause or utterance. Declension is the process by which a word's inflectional ending falls (declines) from a base form (nominative in IE) to an inflected form.

English has this in its pronouns, which decline from a base nominative form to either a genitive (possessive) form or an objective (accusative, indirect, or prepositional) form:

nom | he
obj | him
gen | his

French and Spanish also have this in their pronouns, both possessing a nominative, genitive, accusative (direct object), and dative (indirect object) form. French also possesses a disjunctive form which it uses for emphasis and prepositional objects:

nom | él | il
gen | el suyo | le sien
acc | lo | le
dat | le | lui
disj | - | lui

Latin and Greek have this for all nouns and adjectives, both for Singular and Plural in Latin, and additionally for a dual form (e.g. My [two] hands built...; I saw it with my [two] eyes) in some variants of Ancient Greek (most prominently in Attic and Homeric). Latin has 7 cases - nominative, genitive, dative, accusative, ablative (movement away or separation from a goal, instrument/cause, other random assorted uses), a vocative (for addressing a noun directly as in "et tu Brute"), and occasionally a Locative (static location) in the case of the house, cities, and some small islands (e.g. Lesbos -> Lesbi "On Lesbos"). Greek has 5 cases - nominative, genitive, accusative, dative, and vocative.

Latin has 5 declensional paradigms, determined by how they are inflected in their genitive form:

| I | II | III | IV | V
nom | femina | populus | homō | motus | diēs
gen | feminae | populī | hominis | motūs | diēī
dat | feminae | populō | hominī | motū | diēī
acc | feminam | populum | hominem | motum | diem
abl | feminā | populō | homine | motū | diē
voc | femina | populē | homō | motus | diēs
nom pl | feminae | populī | hominēs | motūs | diēs
gen pl | feminārum | populōrum | hominum | motuum | diērum
dat pl | feminīs | populīs | hominibus | motibus | diēbus
acc pl | feminas | populōs | hominēs | motūs | diēs
abl pl | feminīs | populīs | hominibus | motibus | diēbus
voc pl | feminae | populī | hominēs | motūs | diēs

Because all syntax is indicated via inflections, you can arrange a sentence in any order you want in Latin, moving words around as you want to emphasize the parts of the sentence that are most important. Everything is inferred through the form alone:

Nominative | Genitive | Dative | Accusative | Ablative | Adverb

Hi omnes
lingua, institutis, legibus inter se differunt. Gallos ab Aquitanis Garumna flumen, a Belgis Matrona et Sequana diuidit.

All these people differ among themselves in respect to language, institutions, and laws. The river Garumna [Garonne] divides the Gauls from the Aquitanians, and the rivers Matrona [Marne] and Sequana [Seine] divides them (the Gauls) from the Belgae.

Horum omnium fortissimi sunt Belgae, propterea quod a cultu atque humanitate provinciae longissime absunt, minimeque ad eos mercatores saepe commeant atque ea quae ad effeminandos animos pertinent important proximique sunt Germanis, qui trans Rhenum incolunt, quibuscum continenter bellum gerunt.

Of all these peoples, the bravest are the Belgae, owing to [the fact] that [they] are furthest distant from the culture and refinement of [our] province (Provence), and that merchants least frequently visit them and [they] import those things which tend to effeminate the mind; They also neighbor the Germani, who reside across the Rhenus (Rhine), with whom [they] continuously are waging war.

OKay, rephrasing: do any of the languages you know have a lot of permutations of ordinary words like Hebrew does, and if they do, how many? Like, double-digits?

Personally I don't think there's anything wrong with English. We're long overdue for some spelling reforms,

Yes, what we need is a rashnalized spelling sistim for dhe Ingglish langwij.

(Also, what's your opinion of this?)

Except for the American r: that is an objectively difficult and obscure sound.

First I'm hearing that. You mean that rolling rrrrr is actually the easier sound?
 
Last edited:
OKay, rephrasing: do any of the languages you know have a lot of permutations of ordinary words like Hebrew does, and if they do, how many? Like, double-digits?

Still not really sure what you're asking here. Every Latin noun has at a minimum 6*2=12 different variations. Every (Attic) Greek word has 5*3=15 different variations, depending on its role in the sentence. And each case can fall into any number of dozens of meanings. My Latin textbook lists:

13 different expressions of the Accusative
7 different expressions of the Genitive
11 different expressions of the Dative
18 different expressions of the Ablative

A fun Latin sentence:

malo malo malo malo

"I'd rather be in an apple tree than a naughty boy in adversity"

If we're talking verbs, it gets way more complicated. Every Latin verb has:

3 persons (1st, 2nd, 3rd), 2 number (singular, plural), three tenses (present, past, and future), each of which can be expressed in one of two aspects (perfective, and imperfective), three moods (indicative, subjunctive, and imperative), 5 non-finite forms (infinitive, which can be expressed in past, or present, active or passive; gerund; participial, which can be expressed in past, present or future in active or passive; supine, and gerundive), and two voices (active or passive)

All told that gives a total of:

(6*9)+(6*6)+11+5+6+(4*6)

[indicative non-participle forms]+[subjunctive non-participle forms]+[imperative forms]+[non-finite indeclinable forms]+[gerund/supine forms]+[participle declensions]

or 136 forms for every verb, excepting those which are deponent and so have no passive form, and defective verbs which only conjugate in the third person.

(Also, what's your opinion of this?)

Looks bad.

First I'm hearing that. You mean that rolling rrrrr is actually the easier sound?

The retroflex approximant ([ɻ]) which, at the very least, is what *my* "r" sound is, is extraordinarily rare, existing outside of American English Mandarin Chinese, Pashto, Tamil, and Malayam, only in a handful of nonstandard dialects and some Australian Aboriginal and Indigenous South American languages, and even then only in fits and spurts. It is really quite an uncommon sound.

The dental/alveolar/postalveolar trill ([r]), i.e. the "rolling rrrrr", at least within the context of IE languages exists either as a long-r or as an allophone with the alveolar flap ([ɾ])

It is found in standard forms of: Hungarian, Romanian, Afrikaans, Arabic, Bengali, Czech, Dutch, Finnish, Greek, Hebrew, Hindustani, Italian, Kyrgyz, Latvian, Malay, Polish, Portuguese, Spanish, Tagalog, Welsh, Catalan, and German, among others.

The alveolar flap [ɾ] is found in standard forms of: Russian, Africaans, Basque, Catalan, Danish, English, Greek, Hindustani, Japanese, Korean, Portuguese, Spanish, Tagalog, Turkish, Yiddish, and German

And really one only need think about the shape of the two sounds to understand why the retroflex approximant is rare and the alveolar trill/flap is not. The retroflex approximant involves folding the blade of the tongue down towards the bottom of the mouth and curling it up or outwards, while raising the back of the tongue upwards far enough to produce an effect in the sound quality, but not enough to occlude the airflow enough for a fricative or plosive effect. That kind of tongue gymnastics is hard to produce if you haven't been raised with it.

The alveolar flap/trill is a much easier sound to produce. Positionally it's identical to [t] and [d]. Yes the trill can be tricky initially to get the hang of, but it's something you can pick up after a couple months of practice because it's a sound that's rather similar to other very common sounds, unlike the retroflex which entails an obscure shape that can be nigh-impossible to reproduce if you didn't learn it early enough - which is why English-language learners outside of China have such a difficult time with the sound.
 
Still not really sure what you're asking here. Every Latin noun has at a minimum 6*2=12 different variations. Every (Attic) Greek word has 5*3=15 different variations, depending on its role in the sentence. And each case can fall into any number of dozens of meanings. My Latin textbook lists:

13 different expressions of the Accusative
7 different expressions of the Genitive
11 different expressions of the Dative
18 different expressions of the Ablative

Every Latin verb has:

3 persons (1st, 2nd, 3rd), 2 number (singular, plural), three tenses (present, past, and future), each of which can be expressed in one of two aspects (perfective, and imperfective), three moods (indicative, subjunctive, and imperative), 5 non-finite forms (infinitive, which can be expressed in past, or present, active or passive; gerund; participial, which can be expressed in past, present or future in active or passive; supine, and gerundive), and two voices (active or passive)

All told that gives a total of:

(6*9)+(6*6)+11+5+6+(4*6)

[indicative non-participle forms]+[subjunctive non-participle forms]+[imperative forms]+[non-finite indeclinable forms]+[gerund/supine forms]+[participle declensions]

or 136 forms for every verb, excepting those which are deponent and so have no passive form, and defective verbs which only conjugate in the third person.

I wasn't asking about Latin, just if another popular language forces you to learn twenty different forms of each verb.

Looks bad.

Oh? Well, what proposals do you like?

The alveolar flap/trill is a much easier sound to produce. Positionally it's identical to [t] and [d]. Yes the trill can be tricky initially to get the hang of, but it's something you can pick up after a couple months of practice because it's a sound that's rather similar to other very common sounds, unlike the retroflex which entails an obscure shape that can be nigh-impossible to reproduce if you didn't learn it early enough - which is why English-language learners outside of China have such a difficult time with the sound.

Really? The idea of knowing an incredibly difficult sound from birth sounds pretty cool. But in modern Hebrew, it seems the resh can be both sounds (unless I'm just pronouncing it wrong, which seems unlikely - I'm very proud of my accent).

EDIT: Okay, I may in fact have been mispronouncing it.
 
Last edited:
Spanish has: 3 persons (1st, 2nd, 3rd), 2 number (singular, plural), 5 tenses/aspects in the indicative (present, preterite, imperfect, future, conditional), 3 tenses in the subjunctive (present, imperfect, future), and separate forms for affirmative and negative imperative, in addition to an infinitive, gerund, and a past participle that inflects for gender and number, leaving a total of:

70 possible forms

French has: 3 persons (1st, 2nd, 3rd), 2 number (singular plural), 5 simple tenses/aspects in the indicative (present, imperfect, past historic, future, conditional) - French also has present perfect, pluperfect, past anterior, future perfect, and conditional perfect, but these are all compound tenses that use a variant of avoir/être with the past participle, 2 simple tenses in the subjunctive (present, imperfect; a further two compound subjunctive tenses of past and pluperfect), three forms in the imperative mood, as well as an infinitive, gerund, and past participle form, giving a total of:

48 possible forms, assuming you don't count compound forms, then it rises to 90 possible forms.

Oh and French participles also inflect for gender and number, so 49, and ~132 if you're including compound forms.
 
You're missing the pretérito anterior tense, but if you try to explain the difference between that and pluscuamperfecto to Mouthwash you'll make his head explode.
 
Spanish has: 3 persons (1st, 2nd, 3rd), 2 number (singular, plural), 5 tenses/aspects in the indicative (present, preterite, imperfect, future, conditional), 3 tenses in the subjunctive (present, imperfect, future), and separate forms for affirmative and negative imperative, in addition to an infinitive, gerund, and a past participle that inflects for gender and number, leaving a total of:

70 possible forms

French has: 3 persons (1st, 2nd, 3rd), 2 number (singular plural), 5 simple tenses/aspects in the indicative (present, imperfect, past historic, future, conditional) - French also has present perfect, pluperfect, past anterior, future perfect, and conditional perfect, but these are all compound tenses that use a variant of avoir/être with the past participle, 2 simple tenses in the subjunctive (present, imperfect; a further two compound subjunctive tenses of past and pluperfect), three forms in the imperative mood, as well as an infinitive, gerund, and past participle form, giving a total of:

48 possible forms, assuming you don't count compound forms, then it rises to 90 possible forms.

Oh and French participles also inflect for gender and number, so 49, and ~132 if you're including compound forms.

:wow: I really hope you've misinterpreted what I asked somehow... no language could use that many permutations of a single word.

Let me check: how many forms do English words have?

You're missing the pretérito anterior tense, but if you try to explain the difference between that and pluscuamperfecto to Mouthwash you'll make his head explode.

TIL people into linguistics aren't snobby at all. EDIT: As further evidenced below.
 
no language could use that many permutations of a single word.
Hahahaha, wait until you learn about Bulgarian verbs.

Or Serb(oCroat)ian adjectives.

Or Finno-Ugric nominal declension (you can get over 2k forms of one single noun).
 
tener conjugation.png


Really you just have to memorize the three conjugational paradigms - -ar, -er, and -ir and that handles most of the memorization, though there are several exceptional paradigms such as: stem-changing verbs like morir -> muero; some words with irregular 1st-person singular forms like conocer->conozco, salir->salgo; words that have irregular participial forms like decir->dicho and abrir->abierto; and words that just have their own paradigms like ser, ir, and haber
 
Last edited:
Is there a simple formula for deriving each form? If I learn one form of a word, how many others will I be able to know?
 
Couple things:

1) From a linguistics standpoint your first question is not one that is answerable, nor indeed one which a linguist feels even needs to be answered. All languages are equally good at communicating ideas because all languages exist to communicate ideas. Different grammars are just different ways of tackling the same day-to-day challenges. Things that are inefficient are altered or removed. Elements that aren't addressed by existing elements of the grammar are added to the grammar gradually via innovation.

Linguistics is a discipline which seeks to make no prescriptions about language or how a language "ought" to work. Rather the focus is simply to describe language as it is actually expressed by those speaking it.

2) As said in the post that actually kicked off this discussion: language =/= writing system. A character system like that of Chinese works well for Chinese because of the analytic nature of the language, and because of the difficulties a tonal language like Chinese presents to a phonetic representation via an alphabet or syllabary: any perusal over Latin-alphabet representations of the Chinese language very rapidly make that apparent. Additionally, the fact that Chinese is an analytic language with virtually no inflections makes it very amenable to a pictographically-rooted writing system. Compare this to a heavily inflected language like Latin or English: how to you represent ago ("I do") vs agas ("you do") vs actus ("that which was done") vs agenda ("those things which ought to be done") vs actor ("he who does") vs actrix ("she who does"). It makes far more sense for a language like this to be represented phonetically, and, because of the large amount of vowels (a, e, i, o, u, aa, ee, ii, oo, uu, ae, au, oe) and the large amount of consonant clusters, a syllabary doesn't make a lot of sense. Compare this to, say, Korean or Cherokee, both of which are languages which lend themselves very well to having syllabaries.

Hangul, the Korean writing system is probably the simplest in the world to read, learn, and understand, but it's also a system that was tailor-made for Korean, and really only works because of how Korean works as a language. Latin/Greek/Cyrillic alphabets work extremely well for the Indo-European languages. When your language is built around phoneme-level variations in a word's arrangement of sounds, it follows then that the best way to express your language in writing is through a system in which each symbol corresponds to a phoneme.

Personally I don't think there's anything wrong with English. We're long overdue for some spelling reforms, and the antiquarian in me would love to see þ, æ, etc. brought back, but beyond that there's nothing inherently bad about English that makes it any more or less difficult to learn (in isolation) than any other language. Except for the American r: that is an objectively difficult and obscure sound.
It was a terrific post until you waffled. I might suppose you hinted that English was fine linguistically, but that is problematic diplomatically. Whereas there is a monumental number of Chinese folks but I would guess the symbolic nature would prove challenging to learning it as an adult.

There is something magical and mysterious about the human brain learning language as children. And we lose that capacity or don't care.

I cannot see English as being that relevant 100 years from now, can you? How with the declining population of those with European ancestry?

And it is very intimidating writing a post to a linguist. I cannot imagine all my errors.

A linguist who loves film...what could be more enchanting?
 
Is there a simple formula for deriving each form? If I learn one form of a word, how many others will I be able to know?

Like I said, it mostly comes down to 3 simple paradigms, meaning you really only have to memorize 210 individual forms. There are some irregulars that you have to memorize on their own, but beyond that those 210 forms allow you to conjugate 99% of all verbs. The problem is that the 1% of irregulars is also like, a solid 60-80% of the verbs you most regularly see.

On the one hand with Latin you have to memorize 544 verb forms, not including irregulars, but on the other there are literally only 6 irregular verbs in the whole language.
 
Last edited:
TIL people into linguistics aren't snobby at all. EDIT: As further evidenced below.
Do I have to correct you, or were you using sarcasm?
 
I indeed was.
 
Oh! I wasn't sure, you know.
 
Counties that voted for Hillary produced twice the GDP as those counties that voted for Trump.

You're likely not proving whatever it is you think you're proving.
 
Not everything has to be a thinly veiled punch to the guts, you know.
 
Status
Not open for further replies.
Back
Top Bottom