Lexical decomposition

The lexical decomposition (Greek lexikón "concerning the word"; Latin decompositio "decomposition") (also: lexical decomposition) is a sub-area of linguistic semantics . The main task is to break down individual words or lexemes into their essential meanings, i.e. H. to find the essential aspects of meaning that necessarily contribute to the overall meaning of this expression ( intension ).

Aims of the lexical decomposition

Since the late 1960s, numerous linguists have tried to translate the meanings of all linguistic signs into formal systems, with the main aim of ultimately being able to establish a limited inventory of minimal predicates or semen. Especially the representatives of generative semantics looked for universal systems of meanings and their formalization. This formalization was mainly established by the desire to be able to transfer semantic knowledge to computer systems in order to be able to “calculate” the meanings of all characters in a language. Furthermore, an attempt was made to compile an inventory of semantic atoms that is valid for all languages .

methodology

The meaning of an expression is always determined in relation to another. This is done by forming pairs of opposites, such as B. Girls vs. Woman:

Girl = [+ HUMAN, + FEMALE, -ADULT]
Woman = [+ HUMAN, + FEMALE, + ADULT]

Ferdinand de Saussure already noted that the meaning of an expression depends on the meaning of other linguistic signs. The lexical decomposition does not deal with the development of all the characteristics of an expression, but only with the finding of the characteristics that distinguish them from other expressions according to the principle of the differentiation of meanings (distinctiveness).

application areas

Word fields

A word field can be described as a set of lexemes which have at least one common semantic characteristic and which have paradigmatic meaning relationships with one another. With the help of the lexical decomposition it is necessary to define these common semes or sememes .

The word field ANIMAL has the common features [+ ANIMATED] and [-HUMAN].
In order to further describe parts of this word field, the next smaller distinctive features are used: e.g. B. [+ WINGS] as the most essential sem of a bird.
Possible further differentiations: [+ BLACK] as a special characteristic for ravens.
use.

The terms genus proximum ("the next (higher) genus") and differentia specifica ("the species-forming difference") are very useful for this distinction . In the example above:

genus proximum : "animal"
differentia specifica : "with wings"

A well-known analysis of the word field “University members” illustrates the thesis that hyponyms have at least one characteristic in common with their hyperonym . In this case, the hyperonym “University member”, i. H. all sub-terms have the common characteristic of belonging to the university. The hyponyms only differ from one another through further characteristics:

	college student	Hiwi	assistant	professor	secretary
[UNIVERSITY MEMBER]	+	+	+	+	+
[AUTHORIZED]	-	-	-	+	-
[TEACHER]	-	-	+	+	-
[EMPLOYED]	-	+	+	+	+
[APPRENTICE]	+	+	- / +	-	-
...	...	...	...	...	...

Semantic relations

Semantic relations can be described with the help of semantic characteristics :

Synonyms expressions contain exactly the same semantic characteristics and the same expression.
Heteronymous / incompatible expressions share at least one characteristic with the same expression, but each expression differs from the others in at least one further characteristic.
Hyperonyms contain all the essential characteristics of their hyponyms.
Hyponyms contain all of the essential features of their hyperonym and additional features so that differences can be determined.
Complementary expressions seem to be determined by a single characteristic, whereby one of the two is assigned the value "+" and the other the value "-". For example: "male" = [-FEMALE]; "female" = [+ FEMALE]
Antonyms Expressions can be described with relative features such as [± MINIMAL] and [± MAXIMAL] or [ABOVE / BELOW NORM].

Analysis of verbs

When breaking down verbs, the ability to paraphrase plays a crucial role. The sentence David kills Goliath can be paraphrased by David causing the state to occur that Golitath is dead (= not alive) . This paraphrase divides the verb to kill into three components ( atomic predicates ), with one component each representing the cause, the occurrence of a state and the representation of the final state. These predicates are called CAUSE, BECOME and BE-X (where in our example X = not-ALIVE). In this way, arguments can be assigned to each predicate. CAUSE would therefore have two arguments, namely a causer (x) and the one who is affected by the cause (y); the arguments are shown in brackets:

CAUSE (x, y) or also x CAUSE y
BECOME (x)
BE- [not-ALIVE (x)]

The above sentence would be represented in formal notation as follows:

CAUSE [David, BECOME [BE [not-ALIVE (Goliath]]]
or: [David CAUSE [BECOME [BE [not-ALIVE (Goliath]]]]

Thus, the difference can be made clear when Goliath dies through no fault of your own:

BECOME [BE [not-ALIVE (Goliath)]]

This results in various types of verb actions :

telisch : die ...................... [BECOME [BE [not-ALIVE (x)]]]
causative : kill ................. [y CAUSE [BECOME [BE [not-ALIVE (x)]]]]
stand : similar to .................... [BE [x SIMILAR y]]
Activity : sleep ................... [DO [SLEEP (x)]]

The types of action according to Vendler (1967) can still be defined (using the features [DURATIV / PUNKTUELL, TELISCH / ATELISCH]):

Activity : be silent ........................... [- telisch, + durative]
State : have .................................. [- telisch, + durative]
Achievement : to find ......................... [+ telisch, -durativ / + punctual]
Accomplishment : reading a book ....... [+ telisch, + durativ]

Accomplishments can, according to Dowty (1979), always be broken down into an action and a goal of the action, which can be given as a tripod at the end of the action. As in the following example:

x draws a circle : "x acts in such a way that at the end of the plot the sentence the circle is drawn becomes true" (Dowty 1967)

With overlap with the temporal interpretation: x draws a circle in the minimal interval t shows that the sentence the circle is drawn is false at the beginning of t and true at the end of t.

This purposefulness describes the principle of telicity.

Source

J. Meibauer: Introduction to German linguistics. Metzler, Stuttgart 2002, ISBN 3-476-01851-2 .
HM Muller et al. (Ed.): Arbeitsbuch Linguistik. Schöningh, Paderborn 2002, ISBN 3-506-97007-0 .
D. Alan Cruse et al. (Ed.): Lexicology: an international handbook on the nature and structure of words and vocabulary. Berlin / New York 2001, ISBN 3-11-011308-2 .
Hadumod Bußmann (Ed.): Lexicon of Linguistics. 3rd updated and expanded edition. Kröner, Stuttgart 2002, ISBN 3-520-45203-0 .

literature

Manfred Bierwisch : Formal and lexical semantics. In: Linguistic Reports. 80/82 1982, pp. 3-17.
Manfred Bierwisch: Semantic and conceptual representation of semantic units. In: Rudolf Růžićka & Wolfgang Motsch (eds.): Investigations on semantics. Akademie-Verlag, Berlin 1982, pp. 61–99.
David R. Dowty: Word, meaning and montague grammar. The semantics of verbs and time in generative semantics and Montague's. PTQ, Dordrecht 1979, ISBN 90-277-1008-2 .
Charles J. Fillmore : Lexical entries for verbs. In: Folia Linguistica. 4, 1968, pp. 373-393.
Ray Jackendoff : Semantic structure. Cambridge (MA) 1990, ISBN 0-262-10043-6 .
George Lakoff : Natural logic and lexical decomposition. In: Chicago Linguistic Society. 6, 1970, pp. 340-362.
Anna Wierzbicka : Semantic primitives. Athenäum-Verlag, Frankfurt 1972, ISBN 3-7610-4822-X .
Robert D. Van Valin, Jr. & Randy J. LaPolla: Syntax. Structure, Meaning and Function. Cambridge University Press, Cambridge 1997, ISBN 0-521-49565-2 .