Idea Transcript
CONSTRAINTS ON PRODUCTIVITY The repeated morph constraint (multiple application constraint - Lieber) prevents the same affix to be applied twice in succession. In this form, this constraint is too strong, because there are cases contradicting this restriction: re-re-write in English, Ur-urgrossvater in german, etc. In many cases, the constraint appears to apply to homophonous affixes as well as to identical affixes. For instance, -ly at the end of *month-ly creates A, but cannot usually have the ly that creates Adverbs added to it. In the sense that they create two different parts of speech, these must be seen as two distinct affixes, and yet one constraints the other. See also the case with *sillily. (also *joyfulful, *helplessnessness
Phonological constraints (a) The phonological make-up of the base determines the application of a WF process. Example: English –en forms V from A. Constraint: the stem must end in a stop or a fricative (b) Suprasegmental make-up of the base determines the application of a WF process. Example: The noun-producing suffixal –al (arrival, rebuttal) is added only to V which are stressed on the final syllable. (c) Some cases are are difficult to say whether they are phonological or morphological: Example: Adverbial –ly should not be added to A which already end in –ly. At one stage on the development of English it was a constraint forbidding sequences of homophonous affixes. Now it seems to block the appearance of words like sillily in which the first –ly is not a separate suffix. What was originally a morphological restriction seems to have become a phonological one.
Morphological constraints 3 cases: 1. The base may have to belong to a particular morphologically defined class before affixation can occur. For example, such a morphologically defined class of bases may be a category of gender, or some etymological category (e.g. affixes only attach to foreign bases, etc.) 2. The base may have to show a particular morphological structure before affixation can occur. An affix can be added only to a derived or to an underived base. 3. The base may have to end in a particular affix or not end in a particular affix before affixation can occur. Cases of potentiation (Williams, Aronoff). The application of a particular affix is conditioned by the presence of a different affix.
A more complex case of potentiation is where the affix being added requires not only a particular affix, but a particular structure with that affix., for example, the Dutch – ster requires a structure made up of a suffix attached to a verb. Another case is where a particular type must NOT be present. The suffix –heit which creates abstract nouns from adjectives cannot be added to A derived with a suffix Syntactic constraints The distinction between morphological and syntactic constraints is rather illusory, however, in principle, the morph. constraint concerns the internal structure or inherent class of the word in the base, whereas a syntactic constraint concerns the way the word is used in context Example of SC: Most affixes can be added only to certain word classes (N, V, A). This is what is expressed by Aronoff’s Unitary Base Hypothesis. Another example: A deverbal affixation may be sensitive to the transitivity of the verbal base. For instance, English –able is used on transitive but not intransitive verbs. Therefore *becomable is not possible word. Semantic constraints An affixation is restricted by a particular semantic class, e.g. the German collective suffix -tum is productive only when added to N denoting humans. (Also for example, the derivation by un- restricted to bases with positive meaning – therefore *unill, *unpessimistic unlike unhappy, unoptimistic; or inalienable possession as a condition for –ed compound adjectives: three-legged table, a blue-eyed girl, red-roofed house BUT: *a two-carred man, *a black-shoed lady))
Blocking Morphologists speak of blocking when the unacceptability of a morphologically complex word is not due to the failure to meet some requirement of the relevant pattern of word formation but to the existence of either a synonymous word or a synonymous pattern. In Rainer (1988) it was proposed to refer to the first of these two cases as token blocking and to the second one as type blocking. Token blocking In the case of token blocking, as we have already mentioned, the unacceptability of one morphologically complex word is due to the existence of some synonymous word in the language. The classic example is the unacceptability of stealer as a consequence of the existence of thief. The requirement that a synonymous word exist “in the language” is somewhat imprecise, because what really matters is the existence of an established synonym in the mental lexicon of the speaker or writer in question, not in the language as a social institution (de Saussure’s langue). In fact, it can readily be observed that children, non-native speakers or other people with an insufficient command of the language produce complex words of the stealer type as long as they have not yet internalised the established synonym. Even adult native speakers with a full command of their own language sometimes produce complex words of the stealer type when they momentarily fail to retrieve the established term. This last observation makes it clear that blocking has to do with processing: word formation is only resorted to when no established word is available or if the speaker or writer wants to avoid this established word for some stylistic purpose. The blocking word need not be a simplex as in the case of thief, but may itself be a regular complex word, provided that it is frequent enough to be stored in the speaker’s or writer’s mental lexicon. Frequency is a decisive conditioning factor for token blocking: the higher the frequency of the blocking word, the stronger the blocking effect. In the case of low frequency words, doublets are readily tolerated. What is the status of the blocked word? It is clear that it must be well-formed in principle according to some productive pattern of word formation of the language, otherwise it would not make sense to say that it has been blocked. Nevertheless blocked words should not simply be put on the same footing as possible words, i.e. complex words that are not yet established but readily acceptable to native speakers. Rainerian, for example, a relational adjective derived from the name of the author of these lines, would be such a possible but hitherto non-existent word. And since possible words may serve as input for further derivation, nothing could prevent us from coining Rainerianism should I ever become responsible of establishing some new political, philosophical or other trend. Now, we observe that blocked words, though morphologically well-formed, behave quite differently in this respect from possible words: thief, for example, does not only block synonymous stealer, but also any further conceivable derivation thereof. Stealerless, for example, is just as odd as stealer itself, even though there is no established synonym thiefless to token-block it, and well-formedness is guaranteed by the acceptability of leaderless, teacherless and similar formations. The same holds true for compounds: piano thief vs. *piano stealer, etc. It might thus be useful to replace the traditional dichotomy possible vs. established – or actual – word by the trichotomy potential vs. possible vs. established word. According to this terminology, blocked words would be potential, but
not possible words, in our sense of words which are not only well-formed but also readily acceptable under normal linguistic circumstances. This conceptual distinction allows us to formulate the following general constraint on word formation (in the seventies it would have been called Possible Base Constraint): The bases of patterns of word formation must be possible words; or, put the other way round: merely potential – as opposed to possible – words are excluded as bases of patterns of word formation. Type blocking In the case of type blocking, the unacceptability of a complex word is not due to the blocking force exerted by an established synonymous word, but to the fact that a synonymous pattern takes precedence. Type blocking may thus apply even when no actual blocking word formed according to the rival pattern exists. It will be illustrated here with the rivalry between the synonymous abstract suffixes -heit, -ität and -ie in German after adjectives ending in -il (cf. Rainer 1988: 180-181). [+final stress] [+final stress] & [+learnèd] & /-il/ [+final stress] & [+learnèd] & [+ phil]
-heit -ität -ie
Schema 1 Type blocking with German quality nouns in -heit, -ität and -ie German adjectives with final stress fall into the domain of the suffix -heit, which is fully productive and may even apply to non-native words with final stress, as is shown by akut ‘acute’ → Akutheit ‘acuteness’, galant ‘gallant’ → Galantheit ‘gallantery’, grottesk ‘grotesque’ → Grotteskheit ‘grotesqueness’, etc. The domain of -heit, however, is systematically curtailed by the rival suffix -ität after learnèd bases in -il, which all have final stress, as shown by labil ‘unstable’ → Labilität ‘instability’, steril ‘sterile’ → Sterilität ‘sterility’, etc. It is important to note that even where no quality noun is attested in dictionaries, native speakers will clearly prefer -ität to -heit if the adjective has a learnèd flavour, such as merkantil ‘mercantile’, monofil ‘made of one thread’, etc. Only with some adjectives that have become part of a more colloquial register may one observe occasionally that -heit is also tolerated: debil ‘stupid’ → Debilheit ‘stupidity’, beside established Debilität, skurril ‘droll’ → Skurrilheit ‘drollery’, beside established Skurrilität, etc. We may thus conclude that with learnèd adjectives with final stress -ität effectively type-blocks -heit. Now, it is interesting to observe that the domain of -ität is again curtailed by the even more specific domain of -ie with adjectives ending in the suffix -phil ‘-phile’, which is of Greek origin. Adjectives in -phil should fall into the domain of -ität since they are learnèd, end in -il and have final stress, but nevertheless they consistently take -ie: bibliophil ‘book-loving’ → Bibliophilie ‘bibliophily’, anglophil ‘anglophile (adj.)’ → Anglophilie ‘anglophilia’, etc. The same would also be true of any neologism in -phil. On the Internet, one may only find some scattered formations with -ität, such as Frankophilität ‘francophilia’ or Xenophilität ‘xenophilia’, which are probably best attributed to a lack of acquaintance with the leanrèd vocabulary on the part of the coiners, but nevertheless prove that -phil would fall in the domain of -ität if this suffix were not type-blocked by -ie.