vestiges of gender and case in English

[Igel]

6. Grammatical categories come and grammatical categories go.

Proto-Germanic wasn't so highly inflected. It had a lot of inflection, but it's a piker on the world scene.

For this you need the long view.

English has been conspiratorially moving from inflected to uninflected. If you look at the strong verbs, they've been petering out. Yeah, some verbs like "ride/rided" have gone strong in historical times, "ride/rode". Others only did so dialectally (so I'm a dive/dove speaking, while most others are dive/dived).

The past participle's ("I have ridden" is also giving way to the preterite. So a lot of people are confused when it comes to the forms to say for "I went" vs. "I have gone." That's the trend in English--we've replaced a lot of strong or irregular past participles with the preterite. Had one student do a project where she surveyed students for past part. usage in a variety of sentences and contexts, and in some cases she had 4 or 5 variants. She hadn't noticed them before.

English has hit roadbumps. Most irregularities were regularities. So "sleep/slept" used to be perfectly regular. That <ee> was a long /e/, the <e> in 'slept' was short. Otherwise they had the same quality. Come the Great Vowel Shift and loss of length, suddenly it's irregular. We're gradually fixing this: light/lighted~lit, leap/leaped~lept, dream/dreamed~dreamt. But you can't say "sleep/sleeped~slept." Perhaps in 100 years.

A conspiracy in linguistics is where it looks like language has a goal, is telic. In fact a lot of small, little changes just let that impression happen. So English has moved from reasonably inflected in 1000 BC (when it was Common Germanic) to the present. It'll probably continue to lose inflection. Syntax rules, morphology drools.

The long view, however, is a bit murkier. Indo-European is reconstructed. When you reconstruct, you lose detail, you lose clarity. Irregularities are levelled out by the process of reconstruction. Historical change destroys data, and what's destroyed we can't know. Irregularities tend to be both lost and also introduced--that makes it possible to reconstruct an early new irregularity farther back than it should belong and to utterly miss an irregularity that was completely lost.

Late PIE was inflected and had all those strong verbs, funky ways of relating nouns to verbs, verb forms that were nouns and ways of making nouns into verbs. Early PIE, however, almost certainly lacked most inflections. That's when ablaut--vowel changes--ruled, and so you didn't need all those case endings and all that morphology. It's also very likely that tone wasn't lexical, a property of the word that was often predictable or redundant, but bore meaning, much more like Chinese or Vietnamese. Check out Jay Friedman's dissertation (UCLA, probably late '90s) and other works like that.

This grammaticalization, where there's a cline from fully meaningful words (once upon a time called "autosemantic," having their own meaning) to bleached words (where much of the meaning is lost) to grammatical markers (like auxiliaries) to clitics to suffixes to nothing. This certainly works for IE languages. And for Georgian. Harder to show for other languages without a lot of historical work. I've seen dribs and drabs of this but haven't been paying close attention.

Modals in English show the first two steps--full words to bleached words. Auxiliaries show the first three steps: full verbs to bleached words to free-standing grammatical markers and even clitics. (Our auxiliaries are typically stressless and cliticize to some other word. Judith Klavans did a good job classifying clitics, look up her work, even if it is from the '70s or '80s.)

People balk at the next step: auxiliaries to suffixes. If English had a different sort of clitic it would be easier to grasp. Our clitics lean left, "I've read". If they went the other way, "I v'read" you could see how auxiliary --> clitic --> suffix is a snap. In French the argument's been made that pronouns are clitics and turning into a new conjugational system: "j'li, tuli, illi, nulison, vulize, illiz". Makes Native American agglutinative languages seem completely reasonable, what with object and subject incorporation.

Lithuanian has a rich prefix system for forming verbs. Like "read, re-read, misread", but with more prefixes--through, around, for a while ... We like to think they're all single words, but the word for "no" intrudes. "I misread" would be okay, but instead of "I no misread" you have to say "I mis-no-read." Instead of "I no reread" it's "I re-no-read." You speak the language, and this seems perfectly natural. These prefixes, like the word 'ne' "no", were clitics. They still sort of act like clitics.

Take Latin to Spanish, a typical example, and the future tense. The usual story is something like "Latin had a nifty inflected plural and a nifty inflected perfect." The future had a tense marker -i-, the perfect had what? -av-? (Me, Latin, not a good mix.) When sound change made the endings less clear, more shaky, speakers used "have". This happened the same way in English for the perfect. "I have read the book", a perfect, started out as something like "I have the book read", a resultative; in French you still get agreement with past participles. Or you could say "I have to read the book", in Romance languages that involved putting the "have" verb *after* the main verb. If 'have' is before the verb, it's past; if 'have' is after the verb, it's future.

It's really obvious that's the case in Portuguese. Look at the future and past perfect conjugations there. You can--must, actually--put pronouns *between* the verb and tense markers in the Portuguese future. Those future tense endings were clitics, like the pronouns. And they probably still are clitics in Portuguese, even if they're bound suffixes in the other Romance languages (at least the ones I'm familiar with).

As in French and English, sound change erodes the edges of words, esp. the trailing edges. That's where a lot of grammatical markers show up. So we're always, over millennia, adding markers and then destroying them. We lost the subjunctive that way in English--and French is doing the same. (Not that it's just sound change. It gets messy. Sound change may weaken the morphology, speakers find a way to get around the ambiguity, but in so doing the old morphology becomes meaningless and is preserved in fewer and fewer situations. I'm speaking sloppily here.) BTW, my dialect of English, from a little place SE of Baltimore, colloquially preserved the subjunctive. It drives me crazy to hear "It's important that I'm there by 3". One may as well say, "It were important that I've was there by 3."

English will become less and less inflected. Some will be driven by analogy: "Me thinks he is wise" has as an indirect object "me": "He is wise" it-is thought to me. It makes the thinker more of an experiencer and not the originator of the thought. It's precisely "It occurs to me that he is wise." We interpreted "me" as the subject, and replaced it with I because it was before the verb. (English used to be a verb-second language--whatever else is going on in the sentence, the default position of the verb is after the first stressed element.)

Some loss will be from levelling. "I have gone" vs. "I went" will go the way of most past participles, and "I have went" will almost certainly win. I've already seen supposedly educated people defend its use as an alternative "proper" form, and heard English language learners corrected to "I have went" from "I have go-ed."

Right now English is into syntactic change. Maybe because of influence from other languages, maybe it's just internally driven. "Less" versus "fewer" is biting the dust. "Less and less people like Mozart" instead of "Fewer and fewer people like Mozart." "Every student doesn't understand this" is far from "Not every student understands this," but if you're under 20, or I guess 30 these days, there's a good chance they mean the same thing. "Every student doesn't understand this, but most do" is either okay or blithering nonsense, depending on which side of the grammatical change you're on. (Check out "scope of negation." It's a simple change even if it looks like a complicated one.)

Now, excuse me, we are about to start vectors and elementary kinematics in my high school physics class, and my goal for this year is rationalizing all the problem sets into a single format, with primary, alternative, and review sets for each unit. (Be a linguistics grad student; then teach high-school science!)