Universal

普世的，共相

A term used in linguistics, and especially in generative grammar, referring to a property claimed to be common for all languages, to demonstrate the validity of which is a main goal of linguistic theory. Universal grammar is the term used to identify the main aim of those who hold that the ultimate purpose of linguistics is to specify precisely the possible form of a human grammar – and especially the restrictions on the form such grammars can take. In their broadest sense, then, language universals are equivalent to the general design features of human language identified by some linguists under such headings as duality, creativity, reflexiveness and displacement. In this sense, universals provide a theory of the human language faculty - which is thought to be an important step in the task of understanding human intellectual capacities.

In early generative literature, two main types of universal are recognized. Formal universals are the necessary conditions which have to be imposed on the construction of grammars in order for them to be able to operate. They include such notions as the number of components, types of rules, ordering conventions (e.g. cycles), types of transformations and so on. Substantive universals, on the other hand, are the primitive elements in a grammar, required for the analysis of linguistic data, e.g. NP, VP, [+ grave], [+abstract]. Depending on the component of the grammar in which they occur, universals are referred to as ‘phonological universals’, ‘semantic universals’ (cf. ‘universal semantics’), ‘syntactic universals’, etc. Some of these categories may actually be found in every language, but it is not crucial to the notion of substantive universal that they should be. All that is required is that they be constructs which need to be defined by linguistic theory to enable cross-language generalizations to be made, i.e. they are not terms established for the analysis of just one language, but are capable of general application. The universal base hypothesis in generative linguistics states that all languages can be generated by using the same set of basic rules - thought whether these are seen as rules of the base syntactic component or as a set of semantic formation rules depends on the theory employed.

Other types of linguistic universal have been suggested. Quantitative studies have introduced the notion of statistical universals, i.e. constants of a statistical kind, such as a ratio of use between different structures. Implicational universals are generalized statements of the form ‘if X, the Y’, e.g. if a language has a word-order of a certain type, it will also have a verb structure of a certain type. Absolute universals are properties which all languages share; there are no exceptions. Relative universals are general tendencies in language; there may be principled exceptions.

语言学，特别是生成语法，用来指据称是所有语言所共有的特征，而证明这种共相的有效性正是语言学理论的主要目标。有人认为语言学的最终目标是精确说明人类语法的可能形式，特别是说明这类语法形式上的限制，这样的语法可用普世语法称之。因此在最宽泛的涵义上，语言共相相当于人类语言的设计特征，有的语言学家认为这些特征包括双层性、创造性、自反性和移置性等。从这一意义上讲，共相提供关于人类语言本能的理论，即揭示生物学上必需的那些语言特征；揭示共相被认为是理解人类智能这项任务的重要一步。

生成语法文献中区分两大类共相，即形式共相和实体共相。形式共相是为了使各种语法能够运作而必须施加于语法结构上的条件。这些条件包括语法组成部分的数目，规则的各种类型，序次规约（例如循环），转换的各种类型，等等概念。实体共相则是一步语法的基元成分，是分析语言数据所必需的，例如NP，VP，[+钝音]，[+抽象]等。按其出现的语法组成部分，共相可分别称作“音系共相”，“语义共相”，“句法共相”等。这些范畴中有一些可能实际见于一种语言，但对实体共相而言这一点并不十分重要。需要满足的条件只是，他们是为作出跨语言的概括而需要由语言理论加以定义的概念，换言之，它们不是只为分析一种语言而建立的概念，而是有一般的适用性。生成语言学中的普世基础假说是指，所有语言都可以用同一套基本规则生成—但这些规则是视作基础句法部分的规则还是语义构成规则，这要取决于采纳哪一种理论。

已经提出的语言共相还有其他种类。数量研究提出统计共相的概念，即统计上的不变项，如不同结构的使用比率。蕴涵共相是形式为“如果X，那么Y”的概括性陈述，例如，如果一种语言的词序为某一类型，那么这种语言也有某一类型的动词结构。绝对共相是所有语言都具备的特性，没有例外。相对共相是语言的一般倾向，可能存在有其他规律支配的例外。