Friday, October 18, 2013

Detail #61: Some adverb quirks

Adverbs are a part of languages where conlangers seem not to go into any huge amount of detail. There probably is a bunch of reasons for this, e.g.

  • English having a rather boringly simple system for most adverbs that are formed from adjectives (-ly not giving a lot of potential for interest). This relatively straightforward lack of detail seems to be par for the course in Germanic though (e.g. German and some Scandinavian languages basically having uninflected adjectives double for adverbs, Swedish having adjectives with the neuter marker.) English does do some interesting things, even then: good → well, where well also is partially an adjective (but slightly unusual in that it cannot be an attribute, but can be a complement - *"the well man", but "the man is well"). Another exception is hard, where hardly, of course, is a mostly separate lexeme. This kind of exception can of course be interesting, but I will leave it at that. Coming up with different details than those present in English but of the same kind is left as an exercise for the reader.
  • Adverbs often being subsumed into other existing word classes and phrase classes, e.g. adpositional phrases, adjectives, nouns in different cases. Making verb-like adverbs does not seem to be very common among conlangers, possibly because of them being relatively uncommon in Standard Average European.
  • Adverbs seem to be slightly too far into the bits of a grammar where people
  • don't know what to do with them? 
  • just don't care to do cool things with them? 
  •  don't have an idea that it's possible to do cool things with them? 
  •  Get bored of going into deeper detail or assume no one's going to read that far
So, let's consider adverbs. For now, I will not consider any semantic or pragmatic uses for them - there are ideas I have in those regards that may appear later. What other things can we do to adverbs except something like English does?:
I run quickly
There are two main sources of behaviors I am going to look at, and try to apply their traits to adverbs. These are, (un-?)surprisingly enough, verbs and nouns. I will present a draft of a language's verbal and nominal morphology, and then go on to apply this in different ways to the adverbs.

Nouns

The nouns have three accidents: number, case and definiteness. Not all combinations occur. There are three numbers - singular, plural and partitive. Partitive is always indefinite. Definiteness comes in three forms too - indefinite, known to speaker, known to listener. For this language, the case list will amount to something along the lines of Finnish or Hungarian: 

{nominative, accusative, reflexively possessed accusative, nominative complement, accusative complement, genitive, dative, {{towards, from, at}×{on, in, by}}/{towards,by}, generally away from, instrumental, comitative-with, comitative-to, negative}

As a note, comitative-1, comitative-2 and the instrumental have morphological similarities to the three {towards, to, from, at}-series. The instrumental corresponds to from, and the two others to towards and at. The case I label 'generally away from' also has similarities to the instrumental in form. The negative case takes several rather different roles, but only appears with the partitive number - subjects of negated atelic verbs, non-existing subjects of existential verbs, objects of negative telic verbs, without.

The system is not entirely agglutinating. Although each accident has a default value that is normally expressed by no exponent - e.g. nominative, singular, indefinite - these sometimes do take an exponent - partitive nominative has an explicit nominative marker, as does indefinite comitatives, instrumentals, datives and genitives. Reflexively possessed objects cannot be known to speaker, and are either indefinite or known to listener.

The plural for a rather large group of words is formed by partial reduplication. This partial reduplication also occurs in the partitive for the same nouns. The partitive can signify both plural and singular referents, and this can be shown using verb congruence.

Verbs

The verb has congruence for subjects and direct objects. The congruence is somewhat defective. Partitive-case objects are more likely than others not to be marked on the verb. Some intransitive verbs have object-congruence rather than subject-congruence markers appear for their subjects. Passives entirely lack congruence. Subject congruence distinguishes the three definiteness levels, object congruence merges the two definite kinds.

The verb also marks some kind of TAM-complex. The more complex the set of exponents present (mainly formed agglutinatingly), the greater the likelihood that the verb congruence and/or the tense being deficient. Certain moods lack tense differentiation (although aspect tends to be marked for most of them). Some moods have non-nominative subjects, and for those, the subject congruence invariably settles for zero marking.

As for voices, there are three basic voices: active, passive and oblique. Passive has little to no congruence, and lacks a bunch of moods. Its tense-aspect matrix basically consists of {{past, nonpast× {perfect, imperfect}}.

The oblique voice rather seems to promote non-agent, non-object arguments to subject status. It has {nonpast, past} ×  {imperfect}, thus lacking all perfect meanings. 

On to the adverbs

So, let us have adverbs as somewhat of a wastebasket for words specifying qualities of the VP. To make it interesting, let's make them have properties that make them look like deficient verbs and nouns at the same time. I would probably have two subfamilies of adverbs - the verb-like ones and the noun-like ones, but a clever design could merge these. Probably, some adverbs would be less deficient than others. 

Let's consider a few individual adverbs:
slowly - in this language, this is verb-like, and it takes congruence either with the object or subject (but not both), depending on whether the action is performed slowly on the object, or the agent is performing the action slowly. Essentially, intentional slowness →object congruence, unintentional slowness → subject congruence. The only mood it can mark for is 'intensive', which does not cooccur with subject congruence. With an intransitive verb, object congruence is simply omitted.
harshly - in this language, harshly is noun-like. It is normally partitive instrumental definite (and sometimes in the complement cases), but also can take modal and subj/obj congruence.
weirdly - noun like, but takes object or subject morphology (and the oblique voice). If subject congruence, the "weird" thing is that the subject acted on the object (instead of on something else), if object congruence, the weird thing is that the subject acted, and not that someone else did, and if oblique, the weird thing is some other participant or fact. 
 heavily - noun-like, with no verbal congruence. Partitive instrumental indefinite (and sometimes in the complement cases). 

The complement cases often are used if the adverb describes a subordinate verb or infinite, and then agrees with the subject of that verb (i.e. is it subject or non-subject) 

This is a somewhat unclear description, but I hope it offers some ideas for how to create more complication in the adverbs.

No comments:

Post a Comment