Lexical fields

One of the most fruitful concepts evolved so far in structural semantics is that of the ‘lexical field’, closely associated with Jost Trier and his school. So much has been written of late on this subject that it is unnecessary to go into details. It will be sufficient to recall that lexic­al fields are highly organized and integrated conceptual spheres whose elements mutually delimit each other and derive their significance from the system as a whole. In each field a sphere of experience, concrete or abstract, is analyzed, divided up and classified in a unique way which embodies a scale of values and a peculiar vision of the world. Examples of lexical fields are: the system of colors, the network of family relations; or, among abstract experi­ences, the terms for intellectual qualities, ethical and aesthet­ic values, religious and mystical experiences.

The numerous articles and monographs which have re­cently been published on these problems have all tended to emphasize the differences between these fields in various languages; they have concentrated on what is distinctive and idiosyncratic in them rather than on what they have in common. Yet, beneath all the diversity, there is likely to be an underlying unity which a systematic comparison of these fields would no doubt reveal. Thus we are told of sinking differences between the number and nature of color distinctions: there was no single term for ‘brown’ or ‘gray’ in Latin; Russian has two words for ‘blue’ — sinij, ‘dark blue’ and goluboj ‘sky-blue’; the Navaho have two terms corresponding to ‘black’, one denoting the black of darkness, the other, the black of such objects as coal. Our ‘gray’ and ‘brown’, however, correspond to a single term in their language and likewise ‘blue’ and ‘green’. These differences are highly significant, but it would be equally interesting to know whether there are any elements common to all classifications of colors, any distinctions which have to be expressed everywhere and which could therefore rank as lexical constants.

The same point is even more closely noticeable in another closely organized field which has been extensively studied in various languages: the nomenclature of kinship terms. Take for instance the words for ‘brother’ and ‘sister’. These two concepts seem so fundamental to us that we find it difficult to imagine any language that could do without them. Yet a glance at other idioms will show that they are not in any sense lexical constants. In Hungarian, there was no single term either for ‘brother’ or for ‘sister’ until well into the 19th century; instead, there were, and still are, two pairs of separate words for ‘elder’ and ‘younger brother’ and ‘elder’ and ‘younger sister’. In Malay, on the other hand, there is one collective term for ‘sibling’, which can also mean ‘cousin’. In his report on structural semantics to the Oslo congress of linguists, Professor Hjelmslev summed up the difference between the three solutions in the following diagram:

The three arrangements, though very different, have one thing in common: the general relationship of ‘siblings’ (children of the same parent or parents) is expressible in each of them, either in itself or combined with other criteria. A comparison of the same field in a number of languages would reveal whether this relationship is a semantic universal. It would also show how many ways there are of ‘structuring’ this part of the field and how frequent these various solutions actually are. The same method could then be applied to other sections of the field. Even languages belonging to the same family and culture will sometimes show remarkable discrepancies. Thus there is no single term for ‘grandfather’ or ‘grandmother’ in Swedish: a distinction is made between farfar, the father’s father, and morfar, the mother’s father, and similarly between farmor and mormor. [...]In languages with a different social and cultural background, these discrepancies will be even more marked. In Dravidian, for example, there is an intricate hierarchy of kinship terms based on four sets of distinctions: sex, generation, alliance, and age, of which the third, the only non-biological one, is the most important. [...]

It may be noted in passing that the theory of lexical fields has certain affinities with the Sapir-Whorf hypothesis. Trier and his followers would readily agree with Whorf that each language contains a ‘hidden metaphysics’ and that ‘we dissect nature along lines laid down by our native languages’. There are, however, two important differences between the two schools:

(1) lexical fields have so far been explored mainly in the best-known European languages, whereas Whorf deliberately turned away from ‘Standard Average European’ and concentrated on totally different linguistic systems, notably the American Indian ones;

(2) the theory of lexical fields is focused on vocabulary, while Whorf’s most impressive successes were obtained in the grammatical sphere. It would seem, then, that the two approaches, which have developed independently of each other, could usefully supplement one another, and the time may come when they can be combined into a unified theory.