Home Random Page


CATEGORIES:

BiologyChemistryConstructionCultureEcologyEconomyElectronicsFinanceGeographyHistoryInformaticsLawMathematicsMechanicsMedicineOtherPedagogyPhilosophyPhysicsPolicyPsychologySociologySportTourism






COMPREHENSION CHECK

1. What criteria are used to establish derivational relationship between derivationally unmarked words? Explain the difference between derivationally marked and unmarked words?

2. What is the difference between homophonous and derivationally unmarked words? Think of your own examples.

3. What makes it difficult to define the directionality in zero derivation?

4. Explain how each of the criteria suggested by the author works?


Hans Marchand*

EXPANSION, TRANSPOSITION, AND DERIVATION

1.1. The coining of new words proceeds by way of combining linguistic elements on the basis of a determinant/determinatum relationship called syntagma. It is claimed that there are only two basic categories of word-formation which may be termed Expansion and Derivation. Whether a syntagmatic combination belongs to one or the other category depends on whether the determinatum is an independent morpheme or not. This will be illustrated in the following.

1.2. No syntagma is possible that does not contain at least one independent morpheme. Only this independent morpheme interests us in connection with the problem under discussion. Typifying it be the word boat we will say that boat, when taking part in the formation of a syntagma, may occupy one of two possible places: it will become either the grammatically dominant element, the determination (nucleus, head), or the determinant (satellite, modifier).

The problem is simple in case the word assumes the role of determinatum. Steamboat as compared with boat is a modified, expanded version of boat with its range of usage restricted(see below) so that steamboat, the syntagma, will be found in basically the same semantic contexts as the unexpanded boat. The syntagma steamboat also retains the syntactic primary feature of boat; steamboat belongs to the same word class ‘substantive’ to which boat belongs. An adjective such as colour-blind is an expansion of blind. A person is called colour-blind because he is basically seen as blind though only so with regard to colours. Rewrite as compared with write is basically the verb write with which it is to a great extent exchangeable except for the modification expressed by re-. This does not, however affect the word class of the syntagma, which is that of a verb. The rule for any expansion then will be the following. An expansion is a combination AB which is analysable on the basis ‘B determined by A’ with AB belonging to the same word class and lexical class to which B belongs. All combinations whose determinata are independent morphemes (words) are expansions. Expansions of which both the determinatum and the determinant are words (such as steamboat, blackbird, colour-blind, overdo) are called compounds.

1.3. A further clarification may not be out of place. Semantically speaking, the determinatum represents the element whose range of applicability is limited through the determinant. A steamboat is basically a boat. But whereas boat as an independent unit can be used with reference to an unlimited variety of boats, the applicability of steamboat is limited to those which are powered by strain, excluding those which are not steamboats. We might say that this exclusion in steamboat of non-steamboat things constitutes the determination of boat as performed by the first clement steam, which has therefore been called the determinant. Boat, as the element undergoing a semantic restriction or determination, has been called the determinatum. However, as a syntagma is a grammatical, not a semantic entity, we would say that the terms determinatum and determinant should be defined as grammatical terms. Grammatically speaking, the determinatum is that element of the syntagma which is dominant in that it can stand for the whole syntagma in all positions, as has just been stated in a formula.



1.4. It is important to stress the grammatical character of a syntagma. Semantically speaking, the grammatical determinant is in many cases the part that can stand for the whole combination. This would first apply to copula compounds, coordinative combinations of the type girl friend. Girl may welt fill the place of girl friend, but it has not become the grammatically dominant part. The semantic dominance of the determinant over the determinatum is, however, most in evidence in derivatives containing an appreciative suffix, as in streamlet ‘little stream’. Streamlet is basically a stream though an (emotionally) small one, and could therefore take the place of stream, if semantic considerations were the criterion of substitution. A blackish suit could substitute for black suit as from a purely semantic point of view black has merely been expanded into blackish. But grammatically speaking, black in blackish has lost its independence to -ish just as in blacken it has lost its independence to -en. In either case it is the suffix that dominates grammatically.

2.1. The independent morpheme, as has been stated, may enter a syntagma not as the determinatum but as the determinant. The word boat which is the determinatum of steamboat, is the determinant in boathouse. Colour has the role of determinant in colour-blind, just as cry is the determinant in crybaby. All these combinations are analysable on the basis posited for expansions, with one very important difference. The analysis applies to house, blind and baby respectively. The words, however, which interest us here, the determinants boat, colour, cry are not affected. Boathouse is an expansion of house, colour-blind an expansion of blind, crybaby an expansion of baby. The use of the word in a role other than the determinatum thus does not influence the syntagma, as the only relevant element in it is the determinatum which decides to which grammatical and lexical category the syntagma will belong. As far as the structure of the syntagma is concerned, the result is that any combination containing a word as its determinatum is by definition an expansion. The use of a word as a determinant is therefore less consequential than its use as a determinatum, so long as the determinatum of the combination is likewise a full word. Things are different when the determinatum of the syntagma is a dependent morpheme, as we will see later. But first we must discuss the general problem connected with the use of boat in boathouse, colour in colour-blind, and cry in crybaby. The process is called transposition, a term which calls for an explanation.

Transposition is the use of a word in another than its normal function. Applied to a morphologic syntagma this would mean that a substantive is naturally the determinatum (head, nucleus) of a nominal construction, while an adjective is designed to be its determinant (modifier, satellite). A substantive as determinant, as stone in stone wall, government in government official, colour in colour-blind, is therefore called transposed. Black in blackbird is not a transposition as the adjective naturally modifies a substantive. A verb is by definition a word functioning as the predication of a sentence. Consequently it is not part of a nominal construction. Its use as a determinant in a nominal construction, as in crybaby or G schreibfaul, must therefore be considered a transposition.

2.2. The transposition of a word to the role of determinant in a syntagma where the determinatum is a dependent morpheme is called a derivation. Take the words steamboat and steamer. Formally speaking, both steamboat and steamer contain the word steam in a transposed function, but as in the case of steamboat the determinatum is a word, the whole combination will automatically become an expansion whereas in the case of steamer where the determinatum is a categorising suffix the whole combination automatically becomes a derivation.

2.3. The comprehensive term ‘transposition’ (including derivation), as used by Bally and Sechehaye, refers only to change of word class. There are, however, other cases of transposition. The use of professor in professorship implies not change of word class, as both words are substantives, but change from the semantic class ‘personal substantive’ to ‘abstract, condition-denoting substantive’. We therefore must regard a change from one lexical class into another also as a ‘transposition’ and consequently as a derivation. Change from ‘abstract’ to ‘concrete’, from ‘personal’ to ‘impersonal’ must be considered in the same light as the change from one grammatical word class (part of speech) to another. White becoming whitish stays in the same word class ‘adjective’, but the colour-denoting word white is subsumed under the different semantic head ‘approximating’ expressed by the suffix -ish. Though no change of word class is involved, nobody would doubt the derived character of whitish.

Bally uses the term transposition for various other grammatical phenomena. Campagne in maison de campagne is termed a ‘functional adjective’. This, however, would seem to imply recognition of a status similar to that of the transpositional adjective polar and the presence of a transpositional morpheme ‘adjective’ in campagne. It would involve a series of consequences which are inadmissible, namely the recognition of a transpositional morpheme whenever a word is used in another than its primary function. We therefore will have to make a clear distinction between transposition, which is the general phenomenon, and derivation, which is a problem relevant to word-formation.

2.4. We will speak of derivation only when a word changes its word class (e.g. clean adj to clean vb) or its lexical class (proffessor to professorship). We will not, however, use the term for mere syntactic phenomena as when a substantive functions as the determinant of a group, as government in government official, or stone in stone wall. The use of a substantive as preadjunct is a normal syntactic phenomenon in English, and nothing is there to show that the substantive government here has become an adjective. By contrast, the word governmental in governmental practice will be considered a derivative though admittedly both government in government official and governmental in governmental morale have the same purely syntactic meaning ‘of the government’. But government has remained a substantive whereas governmental, through the adjectival categorizer -al, has become a full adjective, though only a syntactic derivative. The distinction between semantic and syntactic derivatives cannot be gone into here. We will only state that derivation has two facets, a semantic and a syntactic one, the first roughly corresponding to the derivation of semantic units which cannot be ‘generated’ , that is units whose content is not describable solely in syntagmatic terms (as skyscraper, manly, etc.) while syntactic derivatives are satisfactorily explained as renderings of syntactic relations, as writer (of the letter), rendering the S-P relation ‘he wrote (the letter), criminal (court), mirroring the P-prep O relation ‘deals with crime’. Unlike semantic derivatives, syntactic derivatives are generable.

With the exception of E -an (e.g. Roman, American), which seems to form only syntactic derivatives, all suffixes in English, French and German I can think of form both types (cf. E pol-ar bear syn and spectacul-ar sight sem, F végétacion tropic-ale syn and chaleur tropic-ale sem, G der heut-ige Tag syn and eine vernünftige Regelung sem). While mentioning the distinctness of the two types of derivation, it will therefore be advisable to include syntactic derivatives in word-formation, a procedure which is all the more advisable, as many words are partly semantic, partly syntactic.

2.5. We have proposed to include syntactic (transpositional) derivation in the treatment of derivation, because syntactic derivatives (polar, writer, writing) belong to a different word class from their bases. The change of word class is evident in form and syntactic behaviour as one that is irreversible. The transposed words polar, writer, writing cannot be used again in the same word class as pole and write respectively, quite unlike government in government official: there is nothing final in the transposition of government, the word can at any moment be used again as a determinatum in another construction. Its primary character has not been affected by the possibility of its being used as preadjunct. The use of substantives as preadjuncts therefore will not be considered parallel to that of syntactic derived adjectives. A syntactic zero morpheme parallel to the overt morpheme -ar of the type polar cannot be claimed, theprocess is not relevant to derivation.

We might feel tempted to invoke the counter-argument that the existence of an overtly expressed pattern legal/ize, atom/ize induced us to assume a zero morpheme in the case of clean/Øvb and cash/Ø vb, in order to argue for a zero transpositional morpheme in the case of substantives used as preadjuncts. The argument would be perfectly valid if we could show that (government in government official has really become an adjective. No transposed substantive can be called an adjective unless it has received a categorial marker: -al in governmental characterizes the word as an adjective, derived from the substantive government. The case of clean vb transposed from clean adj is clear: as a verb, clean is treated like any primary verb in all positions where it is used, exactly like sing, read, write. Through inflected forms such as he clean-s, he clean-ed, and its use in non-adjectival positions such as he is cleaning the room the verb clean is distinct from the adjective clean. This cannot be said of preadjunctal government as against the independent substantive government.

That government cannot be used in predicative position is of course no argument. Neither can governmental. It is precisely the characteristic of transpositional adjectives that they can only be used as preadjuncts, as they originate in rectional sentences and do nothing but transpose the verbal complement of a rectional sentence into a preposed adjective: polar in polar bear reduces the verbal complement of the sentence ‘the bear lives near the pole’ to a preposed adjective in a nominal construction where bear is the head. The predicate of a rectional sentence cannot be made the predicate of a copula sentence.

There is another case of transposition which shall be discussed as complementary to that of government (official). Any prepositional group can be used as preadjunct, either unchanged (as in he gave a speech after dinnerafterdinner speech), or with the preposition represented by an allomorph not occurring in syntactic use (as in there were riots against foreignersanti-foreign riots), or without the preposition (as in he is an official of the government government official) which is the case discussed above. This means that the phenomenon must be considered a normal syntactic feature in English. The preadjunctal prepositional group is transposed, but it should not be interpreted as a transpositional derivative. The transpositional but non-derivational type anti-Communist (policy) would thus have its counterpart in government (official). Both types are matched by genuine transpositional derivational types: anti-cleric/al (attitude) and government/al (institutions).


Date: 2016-04-22; view: 879


<== previous page | next page ==>
FOR THE ESTABLISHING OF DERIVATIONAL RELATIONSHIP BETWEEN WORDS UNMARKED BY DERIVATIONAL MORPHEMES | WHEN NOUNS SURFACE AS VERBS 1 page
doclecture.net - lectures - 2014-2024 year. Copyright infringement or personal data (0.009 sec.)