【原】“主动词汇”、“被动词汇”，很可能都是错觉（下）（英语教学法原著选读71）

以下内容中的英文原文选自上海外语教育出版社引进出版的《教学法丛书：词汇：描述、习得与教学》一书。

上周一，我们看了《英语学习中所谓“主动词汇”、“被动词汇”，很可能都是错觉（上）——（英语教学法原著选读70）》，里面说到由于测试工具的误差和各种语言、非语言因素，我们直觉中感知到的“主动词汇”、“被动词汇”这一划分实际上很可能并不存在。事实上，几个简单的例子就能证明这一点：英语学习最初的几百个词，包括I、am、what，几乎全部都是学过之后就能应用，从“被动”到“主动”几乎不存在一个转变过程。在下面的选文中，作者Francine Melka提出了对“词汇知识”和教学的一些新观点，节标题为“New views on vocabulary knowledge and vocabulary teaching”。

在这段选文中，Melka提出：在“输入性词汇”到“产出性词汇”（下面选文中分别用Receptive和Productive表示）的转变过程中，有“模仿”（或曰“未吸收复制”）、“理解”、“已吸收复制”、最后是“产出”这几个阶段，而且这几个阶段也并非泾渭分明。Melka认为，“输入性词汇”和“产出性词汇”之间的这多个阶段，可以认为是“词汇知识”水平——或者说是词汇的熟悉程度——的不同。

对于“主动词汇”、“被动词汇”，作者认为这实际上是同一个“语言能力”（language competence）在“理解”和“表达”两方面的不同表现，并不存在截然不同的两个词汇系统。

最后，作者借用Putnam提出的模型，指出词汇教学的核心主要是marker（词汇所属的品类标记，如grapes属于fruit，hydrogen属于element）和stereotype（公认词义，如提起bus我们就想起公交车，提起can我们就想起能力、做得到等）。尽管stereotype不一定准确，甚至可能是错误的（比如can也可能是许可），但这些基本词义长久以来得到公认，必须学会。至于引申义，只要本义掌握得好，一般来说也不会有很大问题。

In analysing the distance between R and P, it appeared that the distance between them, or the plane on which they operate, could be broken up into several stages: starting with imitation or reproduction without assimilation, continuing with comprehension and reproduction with assimilation, and finishing with production. These steps are neither restrictive nor exhaustive, i.e. there may be many more intermediate stages between the two ideal points. This image of numerous stages ‘interrupting’ the R and P line suggests a new way of viewing the distance between R and P. My proposal is that the distance between R and P should be interpreted as degrees of knowledge or degrees of familiarity (as discussed in the first section of this chapter). These degrees are numerous, even infinite, and the passage from one degree to the next is imperceptible, because it has to do with barely perceptible degrees of knowledge of a word.

Moreover, the passage from R to P is not clear and neat: we have seen earlier that even when R is not complete, P already begins because R and P are not ‘watertight’ compartments. In this sense the overlapping and interaction of R and P is great. This suggests that the same knowledge or information is probably used for R as for P. The view of two entities or two systems functioning independently, as presented by Clark (1993) should probably be rejected in favor of visualizing the distance between R and P as a line, a ‘continuum of knowledge’. The line would not necessarily be precisely marked, because of the overlapping of the two notions R and P, but it would stretch gradually from less familiar to most familiar. This way of conceptualization implies that an individual possesses one lexical store which is used either receptively or productively according to needs or circumstances. R and P should be then seen as two manifestations depending on one lexicon. The unique system underlying R and P could be seen as competence (a la Chomsky), and R and P themselves as performance (two sides of the same coin).

The problem of the distance between R and P and of degrees of knowledge is brought back to the more general, perhaps more realistic, and fundamental question of how the unique base of an item should be described and how to determine what is meant by knowing a word. In the vast question of knowledge, the question of the meaning of a word is essential in itself, but also because describing the meaning of a word is prerequisite to the teaching of vocabulary.

The question raised above relates to another: what part of the meaning of a word should be taught? The teaching of vocabulary items necessarily implies a choice in the numerous features related to the meaning of a word. The choice should be made according to a certain priority. I will limit myself to a brief consideration of this semantic problem, using a semantic theory which describes the core of the meaning: this sociolinguistic approach has been developed by Putnam (1975) and introduces the stereotype as a basic unit. The notion of stereotype is a manageable one for practical and pedagogical purposes.

Putnam describes the meaning of a word as a tripartite vector, including the marker, the stereotype and the extension. The marker is a generic feature or classifier (‘liquid’ for water, ‘color’ for red, 'gas’ for argon). The stereotype is a conventional id.ea, which could be inaccurate, speakers spontaneously and intuitively associate with an object, a concept (‘stripes’ for tiger, ‘tasteless’ for water, etc.). The extension is very technical knowledge which a specialist may possess. It is of no interest to us here and will be omitted from our discussion. Marker and stereotype are shared by a linguistic community, i.e, people agree on and accept certain stereotypes even though they know they are false: according specialists, ‘gold’ is not really yellow, but silver like, however, small falsities do not prevent communication between speakers. The stereotype is neither stable nor fixed, since it varies, according to culture, time or individuals’ competence.

This last remark about individuals’ competence is of some interest to us. Wierzbicka (1985:214-18) distinguishes between a concept minimum of an object and a concept maximum. This distinction addresses the variation different speakers have as to the completeness of their concepts. For me, argon is a minimum concept; in other words, it has a weak stereotype: all I know is that argon is a colorless (stereotype) gas (marker). I can use the word even though my knowledge of the object is limited. In the case of concept maximum or rich stereotype (apple, cow, sofa, for example), people in our speech community have a maximum of ideas they share on the basis of their everyday life experiences.

This illustrates the degrees of knowledge as presented previously. Putnam’ marker and stereotype, as non-specialists’ ideas shared by a whole community, should be seen as the core of the meaning of a word. It is this part or chunk of meaning an L2 learner should learn.