Tools for grammatical evolution?

If you're new to these arts, this is the place to ask "stupid" questions and get directions!
Post Reply
Posts: 1
Joined: 29 Nov 2021 06:37
Location: Upstate New York

Tools for grammatical evolution?

Post by AlexChillOut »

Not sure if this is the right place for this, please let me know if it's not.

Are there any tools similar to Index Diachronica for grammar changes? I've been attempting to make a naturalistic language from a proto language, and I have all of the sound changes down, but the grammar is about the same currently, and I'm not exactly sure where to start changing it in a way that would still be naturalistic. The only real thing I've done with grammar is change a few particles into affixes, but I don't think that's exactly enough for what is supposed to be 4000 years between the proto language and the current form of the conlang...
Native: :usa:
Learning: :esp: :deu: :iro: :kor: :jpn: :chn: :fra:
User avatar
Posts: 1184
Joined: 26 May 2012 17:24
Location: Oh, I don't know, I'm probably around here somewhere.

Re: Tools for grammatical evolution?

Post by kiwikami »

The closest thing I've found is the World Lexicon of Grammaticalization, which is unidirectional but has been pretty useful to me in the past (there's a 'source>target' list in the appendices that show what elements seem to like turning into what other elements). Not much in the way of syntactic change, though.

Here's a link: ... calization

If anyone knows of any other resources, I'd be really interested to hear about them, too.
Edit: Substituted a string instrument for a French interjection.

:eng: :mrgreen: | :fra: [:)] | ASL [:S] | :deu: [:|] | :tan: [:(] | :nav: [:'(]
Posts: 2903
Joined: 19 Sep 2011 19:37

Re: Tools for grammatical evolution?

Post by Salmoneus »

It's harder to describe grammatical changes in a chart - they're all more unique, they're gradual, and there's often a lot more of them. It may be helpful to look at some actual languages and their histories. [bearing in mind, however, that the gradual shift from highly inflected Proto-Indo-European to modern generally analytic Standard Average European is not a one-way path: languages don't always become more analytical over time, that's just what's happened in Europe]

In general, when you're thinking about grammatical changes, you're going to be thinking about several different categories of change (though ideally related to one another). These include:

1. Morphological changes
The inflections of nouns/verbs/adjectives/whatever are likely to change over time. Sound changes will attempt to do two things: make paradigms irrelevent by gradually reducing all the inflections to zero; and make paradigms explode by creating more and more and more sub-paradigms as a result of context-dependent sound changes.
[English example: sound changes have gradually wrecked the PIE verbal ablaut system - what started simple had becomee 7 classes (with subclasses) in Proto-Germanic, and by the time of modern English has collapsed completely ino a morass of 'irregular' verbs without clear verb classes at all.]
Indeed, it's not unusual for inflections to be irregular from the beginning - it may not all come from the same source. [eg, if we imagine a future-english developing a locative case, the locative form of some nouns may come from the preposition 'at' being attached, and those of others may come from the prepositions 'on' or 'by', creating multiple noun paradigms without needing sound change]

On the other hand, people don't like either of those outcomes. They attempt to innovate and analogise to keep their language sane. This has various effects:

1a. Changes in markers
As markers fail to 'work' in some instances (either due to exessive irregularity or due to erosion), people can simply introduce new markers, either by innovation, or by analogy from instances (inside the paradigm or outside it) where the markers still work.
[English example 1: to address the problems of strong verbs, past tenses have increasingly been analogised from the simpler weak verbs; those weak past tenses themselves probably began as an innovation (from the suffixation of the verb 'do' as an auxiliary).
English example 2: as the nominal system collapsed through erosion, almost all verbs have adopted the -s plural marker by analogy

1b. Changes in categories
Where an entire category becomes hard to mark, it is possible that rather than innovate new markings people may simply stop bothering to mark it - that category may merge with others.
[English example 1: as the subjunctive has become less and less distinctive, we have largely stopped using it, preferring periphrastic constructions.
English example 2: as noun cases merged in more and more paradigms, they were simply universally merged, even in those remaining paradigms where they might actually still have been left distinct by sound changes.

It's also possible of course for a language to simply innovate an entirely new category! This will generally either be by analogy from some irregularity, or from some non-morphological construction being reanalysed as morphological.

1c. Changes in category semantics
Even where a category remains intact, it's possible for its use to change. Often this may respond to a need elsewhere: if a subjunctive category can't be supported morphophonologically, it may simply be replaced wholesale by an optative, for example. Other times this occurs with little motivation, simply shifting a grammatical burden from periphrasis to morphology.

1d. Changes in paradigms
Where paradigms are no longer distinct, they may be collapsed together. [eg English nouns]. Alternatively, where they are increasingly heterogenous, they may be divided into different paradigms. Paradigms may even be collapsed together when fully distinct, usually because one paradigm is rare and non-productive.

Grammatical changes are often slow and piecemeal; the details are often complex (particularly around innovations and analogies - linguists can have long debates about where new ideas came from); but there are also often big-picture general developments.

To take the known history of English verbs, for example:
- early PIE had an ad hoc derivational aspect system and two paradigms for person marking, as well as complex ablaut processes in its core verbs
- by late PIE, the ad hoc derivational aspect system had been converted into a systematic - but complex/irregular aspect-tense system. The difficulties of the original 'athematic' verbs were being sidestepped by an easier 'thematic' paradigm, which particularly included denominative verbs. However, verbal systems were still complicated - verbs have to be described through a 'principle part' system instead of regular conjugations
- by Proto-Germanic, the verbs have been (almost) entirely regularised. Moreover, the system has been simplified - the subjunctive has been lost (the old optative becomes a subjunctive), and tense-aspect have been reduced to a single pair of tenses, present and past. [leaving a small number of weird verbs - 'will' (from the optative) and the preterite-presents]. Secondary verbs are given innovative new past tense endings
- by Old Northwest Germanic, the dual is gone (or almost), there have been futher mergers of forms, and the passive forms have been lost
- since Old Germanic, in English the weak verb classes have all be analogised down into one, while sound change has pulled apart the strong verbs, leading to most of them becoming weak, and the remainder being considered irregular, while the preterite-presents have become entirely defective

So really, you could sum up the last 6,000 years or so of English verbs as:
- a system with few or no inflectional distinctions was hammered into a regular system of conjugation with many categories [pre-PIE through to post-PIE]
- this sytem was complicated, so has undergone radical simplification both in number of paradigms (despite the efforts of sound change) and in number of categories [pre-Proto-Germanic through to Modern English]


Aside from morphological changes, you then also have:

2. Typological changes
These are a bit harder to characterise, as they're mostly less objective (aside from things like alignment). But they represent differences in the general ways languages go about doing things. To take the English example again: from PIE to PGmc, and from PGmc to ME, there has been a gradual shift away from highly synthetic forms to analytic forms, and also from noun-y forms to verb-y forms. So, for instance, PIE verbal noun forms were greatly simplified by the time of PGmc, and even more so since then, and have less and less been considered nominal. PGmc still had at least two, if not three case-marked infinitive forms, whereas now we have only one infinitive, not marked for case at all (and our gerund is relatively rare). Instead, we prefer to use lots of auxiliary verbs.

3. Syntactic changes
Word order can change (which can trigger typological and morphological change). PIE was mostly SOV with postpositions, which over the history of English has developed into SVO with prepositions. The precise route of this has led to things our two layers of prepositional verbs (older "overlook" style verbs from the SOV era, and newer "look over" style verbs from the SVO era), and do-support in questions (verb fronting for questions combined with fixed verb-after-subject positioning, resolved by fronting of a dummy auxiliary verb), and so on.

In regard to both these types of changes, you can see how the big-picture PIE>English story is easy enough to describe (denominalisation/desyntheticisation, shift of the verb forward to follow the subject), but the details may be very complicated (which is part of what makes English different from related languages!)
Post Reply