The Underlying Facts of Language: Evasion and Vocables

R.E. jennings is a philosopher who is intrigued by some of the ways we use language. He believes that misunderstandings in the scope arrangement of some functionalized relational vocabulary (such as or, and, but, until, unless, while, without, etc.) go unnoticed and lead to new uses of this vocabulary. Jennings refers to these occurrences as scope evasions.

The length of the following quote, by Jennings, is understandable in view of the fact that the ideas it contains were the catalyst for this research.

The construction of an explanatory theory [32] raises thorny problems of vocabulary and representation. Space does not permit a detailed justification of the adoption of biological language or the use of descriptors like species or inferential effects. The theory must accommodate many developments that are omitted from this discussion. Suffice it to say that a species is understood as a union of morphologically identifiable populations temporally ordered by an engendering relationship (earlier populations engendering later ones) and that effects are understood as neurophysiological effects occasioned by uses and available to later processes of speech production.

The uses of the word or provide a good example of the need for such a theory, especially as the assumption that or is always disjunctive has befuddled almost all twentieth-century logic text authors and infected the understanding of several generations of theorists brought up on their teachings. Contrast the sentence ``It will be useful or it will be harmful" in which or is disjunctive, with the sentence ``It could be useful or it could be harmful" in which or could be replaced by a semi-colon or even an and or but without serious change of meaning. Or has many other non-disjunctive uses distinct from this one. English also affords many examples of connective vocabulary having among their various uses or/and pairs of meanings and and/if pairs. For an example of the latter, consider the uses of without in ``She will die without fear" and ``She will die without help". To see how this dualization comes about, consider the word unless, a contraction of the expression on [a condition] less than that. The longer expression ought to mean something like and not; however unless means something more like if not. The reason for this seems to be that the gradually contracting construction that eventuated in unless always occurred within the scope of some negative sentence element. But an increasing segment of the linguistic population misinterpreted the scope of the negative element, reading ``not (A unless B)" as ``(not A) unless B". Since the two population understood the whole construction in the same way, the unless had to be construed as an or by the one while being construed as an and not by the other. Eventually unless migrated to non-negative environments carrying with it the new meaning. In this case the older meaning died out. In many other cases, both meanings survived, generally with the new meaning marked against ambiguity. The development appears to have been a case of syntactic metanalysis, a phenomenon hitherto noted [Jes22] only at the level of word-formation, which has yielded an umpire from a numpire, an apron from a naperon, a nickname from an ickname and so on.

At the syntactic level, the structural characteristics of this kind of mutational development are repeated in all of the many other cases of connective dualization by scope evasion, including cases where negation is not implicated. Five stages can be distinguished:

1.

. Stage one: The initiating scope misapprehension (by members of population B) takes place, typically involving negative and modal sentence elements (in the speech of population A). The resulting combination of approximately correct (or at least undetectably incorrect) apprehension of satisfaction conditions on the one hand and undetected incorrect processing of syntax on the other forces an incorrect (or at least novel) apprehension of connective meaning. This is the misapprehension stage.

2.

. Stage two: The novel meaning in the B use of the connective is nurtured by the fact of its uses coinciding in satisfaction conditions with A uses. Since at this stage the new meaning may be thought of as lying hidden beneath the old, it is called the succubinal stage.

3.

. Stage three: The connective in its novel meaning appears in environments in which it is not hidden by the older one. But there is a sufficient B population that the use is not corrected, and members of the A population read it as satisfaction conditions seem to dictate. This is the migratory stage.

4.

. Stage four: The B meaning is sufficiently established that when the connective occurs in the environments that spawned the new meaning, it is ambiguous as between A and B readings. This is the ambiguity stage.

5.

. Stage five: The ambiguity is removed by a marking of the new meaning either prosodically or by the addition of elements that cue the B reading where it is wanted. This is the marking stage. Thus we have for all (compare ``I am steady, for I am sober" with ``For all I am sober, I am none too steady"); just any (compare ``If anyone can join, I can" with ``If just anyone can join, then I don't want to"); even if (compare ``If the Queen asks me, I'll do it" with ``I won't do it even if the Queen asks me") and so on.

-Jennings, 1998, p.1-

The specific idea that Jennings' [48] work inspired is how the loose use of some vocabulary leads to unnoticed mistakes, mistakes that have observable consequences in the syntax of language. From this idea has stemmed another; the imprecise way in which we use this vocabulary hints at a dynamics that has profound repercussions on the shared aspect of language. The first question we ask is: Why is it so difficult to enunciate a straight-forward semantic theory for connective vocabulary? Can this question be asked about all vocables of the language?

By vocables we mean all vocabulary of a language and all of its ancestors distinguishable and different from the sort of noises that are usually associated with non-human primates, such as primate calls. Not all vocables are as semantically difficult as connective vocabulary or become functionalized, but the apparent erosion of semantic space is evident in all diachronic accounts.

Jennings' work has also involved the tracking of change in lexical vocabulary. One of the most dramatic changes that lexical vocabulary can undergo is functionalization. Functionalization is the process that generates functional vocabulary such as connective and auxiliary vocabulary. We restrict our definition of functional vocabulary to include vocables which come to have a new syntactic role that is almost entirely divorced from their ancestral uses. Have is such a word: In its original use, have is found in the context of ownership: I have a dog. But in its functional use, have has lost its connection to ownership and has adopted an auxiliary role in the composition of past tense form of verbs, as in, I have done it already. Notice, as well, that, in the spoken form, the change in structural role has been accompanied with a morphological truncation, from have to the contraction `ve. So one would say: I've done it.

From observing many such examples, mostly in English but also in other languages, (such as the Indo-European family of languages, sign languages, pidgin and creole languages, child and adult forms of language, speech pathologies) a pattern of behavior slowly emerges. The pattern is one of change and transformation in the use of all vocables. These changes can be historically tracked and we can give a formal account of the dynamics of these changes. Our formal account involves a theoretical description of a system of linguistics that propagates through a population. The linguistic dynamics, from which we derive rules that define our language, explain the examples of language change that surround us.