Estimates of the relative productiveness of one or another type of word formation are subject to many variables and consequently uncertainties. Not least among those is establishing the correct etymology of a word. For example, unconscious 'that part of the mind not available to introspection, which nevertheless affects behaviour' might reasonably be thought to be either a shift of use from the adjective or a clipping of the collocation unconscious mind, or even a reformation with the prefix un-. The OEDs first citation, dated 1884, is from Mark Pattison's Memoirs: 'I cannot help observing the remarkable force with which the Unconscious - das Unbewusste - vindicated its power.' That citation suggests that the English word is a caique on German and therefore a borrowing. Such uncertainty is far from unusual.
Any single estimate of the frequency of various types of word formation will also be skewed because of the sample of words examined, the etymological categories used, and the way the categories are applied to the sample. Consequendly, different estimates are seldom fully comparable, but using several estimates rather than one has the advantage of one estimate's cancelling out the idiosyncrasies of another. Below are the percentages for six estimates:
(1) 1,000 words from The Barnhart Dictionary of New English since 1963 (Barnhart, Steinmetz & Barnhart 1973; reported by Algeo 1980), a sample of about one-fifth of the words in that dictionary;
(2) 1,220 words in The Longmnn Register of New Words (Ayto 1989), all the words in that dictionary;
(3) 393 words from The Oxford English Dictionary, 2nd edn. (OED 1989), a sample consisting of the first form or sense on each of the 1,019 pages of volume 1 (A-Ba^ouki), provided that form or sense had an earliest citation date of 1776 or later;
(4) about 500 words beginning with the letter^4 and first attested after 1900, taken from NEWS (New English Words Series), a collection of some 5,000 words not in the OED or OEDS, as analysed by John Simpson (1988);
(5) 2688 words from volumes 1-4 of The Barnhart Dictionary Companion (Barnhart & Barnhart 1982-) as analysed etymologi-cally by David K. Barnhart (1987, 53-69);
(6) 16,570 words analysed by Garland Cannon (1987), consisting of 4,927 words in Barnhart, Steinmetz & Barnhart (1973), 4,536 words in Barnhart, Steinmetz & Barnhart (1980), and 7,107 words in the addenda of the 1981 printing of Webster's Third (1961).
The first three samples are the smallest, but were analysed by the same set of criteria. The sixth and largest sample includes all the words analysed in the first sample, but because of the size of the sixth sample, that duplication does not seriously affect the results. The percentages of etymological types in these six samples are as follows:
Barnhart Longmnn OED2 NEWS BDC Cannon
Creations 0.0 0.0 0.3 0.0 0.0 0.6
Shifts 14.2 19.4 23.4 30.8 9.6 19.7
Shortenings 9.7 10.0 1.8 17.5 9.7 17.1
Composites 63.9 54.3 52.2 37.6 73.5 53.8
(Compounds) (29.8) (36.3) (19.8) (12.0) (57.6) (29.6)
(Affixations) (34.1) (18.0) (32.3) (25.6) (15.9) (24.2)
Blends 4.8 9.8 3.3 1.1 0.5 1.0
Loanwords 6.9 4.3 18.8 6.9 6.2 7.5
Unknown 0.5 2.2 0.3 0.0 0.5 0.4
Others 6.1
The 'others' category exists for the NEWS corpus because 6.1 per cent of its words were not reported for any of the major categories, perhaps because of discrepancies between the simple taxonomy used here and the seventy word-formation categories used for NEWS. Since the approximately 30 words represented by that 6.1 per cent should be distributed among the major etymological categories, the figures for that corpus would change slightly.
Several large variations in percentages between the corpuses are explicable.
Shifts, the NEWS corpus is above and the two Barnhart corpuses are below average in their number of shifts. Shifts of meaning and grammatical use are more likely to be recorded in the OED than in dictionaries of new words because they are less obvious and of less interest to many readers of the latter (Simpson 1988: 151). Shifts of sense and grammatical category are undoubtedly more frequent than most lists of new words would suggest.
Shortenings: shortened forms are markedly higher in percentage in the NEWS and Cannon corpuses, and lower in the OED2. It is particularly striking that the highest percentage of shortenings is in the new words collected for the OED (NEWS) and the lowest percentage in the OED2 itself. To some extent that discrepancy may reflect the fact that clipping, acronyms, and alphabetisms have become increasingly fashionable in recent years. But much of the discrepancy doubtless results from the OED% practice of running in acronyms and alphabetisms under the initial letter of the alphabet; they are therefore under-represented in a sample based on distribution through the pages of the book. The other corpuses treat acronyms and alphabetisms as main entries in normal letter-by-letter alphabetised order. Nearly 7 per cent of Cannon's shortenings are words that might be alternatively analysed in other ways. Composites: compounds are markedly more numerous in The Barnhart Dictionary Companion for several reasons: idioms like keep one's feet to the fire, entered in it, have been counted as compounds here; forms like telework, teleworker, teleworking are listed as independent compounds in its etymological lists, whereas the analysis of other corpuses would treat them as related to one another by affixation or backformation. The OEDs practice of listing compounds as run-in rather than main entries has as a consequence their undercounting in a sample based on distribution through its pages. However, the exceptionally low percentage of compounds in the NEWS corpus is puzzling. Part of the explanation for it may be that the percentage of compounds reported was only for nouns of three patterns (n+n, a+n, v+n) and adjectives of two patterns (n+a, a+n). A goodly proportion of the 6.1 per cent of 'other' words may be compounds of other kinds.
On the other hand, the OED enters even predictable affixed forms with greater fidelity than most dictionaries of neologisms, and therefore has a larger share of them. The Barnhart Dictionary of New English also pays more attention to affixation than average, whereas The Longmnn Registered The Barnhart Dictionary Companion pay less, but such fluctuations may reflect only the lexicographers' focus.
Blends: blends are over-represented in The Longmnn Register perhaps because it includes a good many voguish and nonce forms, which favour the process of blending.
Loanwords: borrowing appears more often in the OED perhaps because of the longer chronological range of the sample taken from it (more than 200 years); the other corpuses report new words from a recent twenty-five year period. There is some reason to suppose that borrowing is less influential as a kind of word derivation now than it was formerly. The OED also, however, prefers to cite Latin and Greek etyma when the formation of a word may be accounted for by native morphemes ultimately of classical origin, and that preference exaggerates its percentage of loanwords.
For example, the OED derives adscription in the sense 'ascription' (first attested in 1857) from Latin as an adaptation of adscnptionem. However, there is nothing in the available evidence to suggest that the word was taken direcdy from Latin, rather than formed from the prefix ad - and the stem scription, both of very long standing in English. Many of the words etymologized by the OED as borrowings may more properly be native formations from morphemes ultimately of foreign origin. In some cases, it is likely that both processes operated simultaneously, which would in effect make blends of the words derived from both a classical etymon (adscrip-tionem) and English formatives (ad - and - scription). Without relying on specific percentages, it seems clear that overwhelmingly the major source for new words in English is their composition from morphemes already present in the language, by compounding and affixation. A distant, but still clearly, secondary source is the shifting of old words to new senses and uses. Shortening, borrowing, and blending are relatively minor sources for neologisms. The creation of words independently of any etyma is insignificant.
Buy custom Literature essay, Literature term paper, Literature research paper.
