LB 98 伊津 IV. PROBLEM ARISING FROM RATE AND ORDERING

IV. PROBLEM ARISING FROM RATE AND ORDERING
Karl Lashley(1951) was the first to recognize clearly the problems raised by the fast rate of movements and the ordering of motor events; the solution presented here is essentially similar to this.
Karl Lashley(1951)第一個清楚地認識到由於動作快速的頻率和運動事作的排序所引起的問題；這裡提出的解決方案本質與此相似。We have postulated some automatisms that are responsible for the fast sequence of movements in speech (as well as many other type of motor behavior). What might the nature of such automatism be? Could it be an associative sequential process? Disregarding for the time being our inability to define association neurophysiologically, from a logical point of view, let us see whether temporal association might account for the facts. The formal characteristic of the associational automatism to be considered is that events occur in chains. For instance, a stimulus is followed by a response; the response then acts as a new stimulus (perhaps because the subject has heard himself say something or feels his own muscles move) which in turn elicits another response; this again becomes a stimulus which is followed by a response, and thus a chain reaction is produced.
我們已假設某些自動作用是說話中快速的連串動作(也是其他類型的運動的原因)的原因。這樣的自動作用的本質為何?它可能是組合性的連續過程嗎?暫且不管我們不能就神經生理學上來定義組合(association), 從一個邏輯的觀點，讓我們看看是否時間上的組合可以解釋以上的事實。我們要考慮的組合性自動作用的形式特徵是事件的發生像是一連串。舉例來說，一個刺激之後跟著一個反應，這個反應接著又作為一個新的刺激(或許是因為受試者已經聽到自己說了什麼或者感覺自己的肌肉有動作)，而這個新的刺激接著又引發另一個反應；這個新的反應又再度成為一個刺激，而後又跟一個反應，於是一個連鎖反應就發生了。
Generally speaking, any one event is triggered by one or more events that had preceded it. For instance, in the application of this principle to phonology, one phoneme is thought to heighten the probability of producing a given other one (by virtue of earlier temporal contiguity in the experience of the organism); but once a phoneme has been produced, it cannot be modified, logically, by phonemes yet to come. Thus this model (let us call it the sequential chain model) may account for modifications or occurrences ”down stream,” namely as consequences of earlier articulatory or phonological events; however, it is unable to account for the phenomenon of anticipation. Nevertheless, articulatory anticipation is a reality as indicated by the pathological example cited previously, and there are cogent physiological reasons that force us to adopt a model that can account as easily for anticipation in articulatory output as modification due to earlier occurrences.一般而言，任一事件都是由該事件之前的一項或者更多的事件所觸發。舉例來說，這項原則如應用於音韻學中，一個音位會被認為提高發出另一個特定音位的機率(by virtue of earlier temporal contiguity in the experience of the organism)；然而一旦一個音位已被發出，邏輯上來說，它就不能被尚未發出的音位調整。因此這個模型(讓我們稱之為連續鏈模型)可以解釋下游的調整和事件，也就是作為較早的發音或音韻事件的結果；然而，它並不能解釋預期同化(anticipation)的現象。不過，發音的預期同化，如同先前引用的病理案例指出，是一個事實，而且有確切的生理學理由促使我們採用一個模型，它不但可以輕易地解釋發音產出的預期同化也可以解釋基於較早事件而來的調整。The reality of anticipation is best seen in the fact that a given initial sound, say /k/ has different acoustic qualities (in English) if followed by an /i/ than when followed by an /u/. Chomsky (1957) has also shown that a sequential chain model is incapable of accounting for almost any aspect of syntax (see Chapter Seven and Appendix), but here we are more concerned with physiological reasons for rejecting the sequential chain model.
預期同化(anticipation)的現實於以下的事實可以最清楚的看出來，一個給定的起始音，像是/k/，其後的音是/i/或/u/會有不同的聲學性質(在英語的情形中)。Chomsky (1957)也已指出一個連續鏈狀的模型幾乎無法解釋句法的任何方面，但是，我們在此更關注的是能夠駁斥連續鏈狀模型的生理學理由。