next up previous
Next: What we have demonstrated Up: The Experiment Previous: The simulation

Results and interpretation

Preliminary results show that the model is consistent with observational data.

Figure 6.10 illustrates the vocable grid. We observe that, in its settled state, the lower portion of the grid is almost entirely dark while the upper portion is almost entirely light. The middle of the grid show a downward increase in clustering of dark constituents.

  
Figure 6.10: The spin grid illustrates the effects of activity in the system in the context of degrees of attenuation in vocables. Notice that the lower portion of the graph shows a greater distribution of dark, indicating that functional instances are in a majority. The middle portion illustrates some clustering which may indicate occasional functional instances however, unstable.
\includegraphics[scale=0.70]{spingrid2}

We expect the A portion of the grid to remain mostly lexical (light) because constituents in that region have no attenuation. However they are subject to the bias field that keep these constituents in a lexical state. This is comparable to the very early uses of vocables that are entrenched in perceptual cues. In fact, some vocables do not leave that stage. (e.g. some nouns and verbs)

In the B area of the grid we observe small sparse clusters (dark areas). This is equivalent to the appearance of aberrant occasions of functional uses. These are ultimately not sustainable because the critical point has not been reached for those constituents. In linguistic interactions these occasions of functional vocabulary will not survive because they may not generate enough specific effects so they will be ignored or perceived as mistakes and corrected.

In the C area of the grid, the number of clusters augment while the majority of constituents still remain in a lexical state. This could indicate rare but stable pockets of functional uses. Again, because the critical point for these constituents has not been reached either, these functional uses are not stable. This may be equivalent to dual values in the use of some relational vocabulary. Often one of these uses will eventually disappear or be tagged for disambiguation. Consider the uses of thanks to. In the construction;

Thanks to Sue, we have met our goals. This lexical use of thanks to is commonly used.

Thanks to the weather, we had a good run. In this example, the slightly less commun use of thanks to is equivalent to the use of because.

Because of the weather, we had a good run.

In the D area of the grid, most constituents are in a functional state. The critical point for these constituents to change state has been reached. This is a stable state for constituents. This is when lexical uses of a vocable become sparse and isolated and mostly functional instances are in use. It may be that some lexical instances persist in the language but they do not generate specific effects or that later lexical instances have strayed considerably from early ones. Consider the use of if. Its Anglo-Saxon roots, giefan or gyfan - related to the current uses of given that - has dissapeared from Modern-English but can still be found in written litterature. However, these written occasions will only generate specific effects with certain linguists, but not in the general population.

We have simulated the lifespan of a meta-vocable in a simple simulation as it changes because of linguistic activity. It is unlikely that any vocable in the language follows an identical path as the one we have modeled for the meta-vocable. However, we have illustrated, with a few examples, that several types of vocables display behavior similar to the ones we have simulated.


next up previous
Next: What we have demonstrated Up: The Experiment Previous: The simulation
Thalie Prevost
2003-12-24