The kinds of changes that we are tracking are the kinds of changes that we have described with our example have. We showed how its lexical use of possession, in some instances, is shed, to give rise to a functional auxiliary role, as in I've gone shopping. This is considered a transcategorial change. Two more such changes are the adverbialization of verbs (I have gone RUNNING) and the prepositional use of adjectives (The shop is DOWN the street). The functionalization of vocables occurs mostly with relational vocabulary, such as even if, without, while, over, and so on. Their lexical ancestors often bore spatial or temporal uses. These are usually shed by their functional counterparts. There are a few exceptions, however, such as in fact whose ancestor is a noun.
In his paper, D.K. Johnston [34] traces the natural history of in fact through its first uses as a noun. It started out resembling the use of in action as gracious in fact, not in word[2]. It is now used as based on observed events as in the fact that this happened.... It is also used functionally, as in In fact, the chance of recovering the stolen items is unlikely. Despite its original use and role as a noun, in fact has been extended to adverbial uses and is rarely used as it originally was in its early uses. This is an uncommon example: most nouns do not develop functional uses. However, it demonstrates that there are no type of vocable immune to functionalization given proper circumstances, circumstances that we will later describe.
We have said that a common context can emerge from a neighbour interaction negotiation. We have also said that individual metaphors are in a form of competition that resolves itself with a perceived overlap of features. Now imagine that such a transaction is successively repeated over time by every individual in a population. As vocables are negotiated, the specific effects associated with vocables will become eroded because the empathy effects, of the kinds reported by Rizzolatti, triggered in different individuals will increasingly become difficult to correlate between them. The increasing discrepancy between the individual use of vocables leads to a loss of common features which in turn leads to attenuation.
One consequence of attenuation is well illustrated by Jennings example of scope misapprehension [48]. These are the mistakes that go unnoticed in the use of vocables that eventually allow them to take on new syntactic roles. We may ask: What is the advantage in the functionalization of lexical vocabulary or in the grammaticalization of linguistic forms? It may be related to Bichakjian's claim of efficiency. Like word order, functionalization allows for flexibility and complexity of expression. We can imagine that, as human activities require more precise coordination, linguisticity will also have to retain or augment a certain level of precision across an increasingly large population that shares it.
The fact is, there are vocables in the language that are semantically difficult and their use is not lexical. And, or and but are such vocables. As Jennings explains, it is not entirely clear what the semantic role of logicalized vocabulary is - vocabulary such as or defined truth functionally - but its role seems purely syntactic. We also know that a diachronic account of these vocables shows that functional vocabulary descends from lexical vocabulary and that, in the course of its history, some lexical vocabulary will undergo a transcategorial change of the functionalized kind.
Our theory suggests that there is a critical point at which attenuated lexical vocabulary will no longer generate specific effects. It has become too attenuated. A consequence is the disappearance of this vocabulary. If it has no specific effects, why use it? Another is a new syntactic role, accompanied by a morphological change.