Efficiency in Linguistic Change

by Otto Jespersen (1941)

1. Evolution and Progress

1.1. In my youth I was, like so many of my contemporaries, under the spell of what Sapir (Language 130) somewhat unjustly termed 'the evolutionary prejudice', Darwin's and Spencer's theories. Into the latter I was first initiated through the philosophical lectures of Professor S. Heegaard during my freshman's year (1877-78). It stamped the whole of my intellectual outlook, and when I first began a serious study of philology I tried to apply this theory to the history of language, though I soon saw that Spencer's famous formulas of evolution (integration, heterogeneity, definiteness) could not be strictly and dogmatically applicable in language. I took "Progress in Language" to mean something totally different from what Spencer spoke of in the linguistic paragraphs of his essay "Progress, its Law and Cause" (Essays, vol. 1): he there speaks exclusively of greater and greater heterogeneity - and increasing number of parts of speech, of words to express the most varied ideas, of languages and dialects produced by the splitting up of one uniform language. I took progress in the more popular sense of advance in usefulness, which Spencer here totally neglects.

Still I had some points of contact with Herbert Spencer. I had early been impressed by his essay on "the Philosophy of Style" (in Essays, vol. 2). In this he says that the best style is that which pays most regard to the economy of the recipient's attention. "Other things equal, the force of all verbal forms and arrangements is great, in proportion as the time and mental effort they demand from the recipient is small." Again, "there is an expenditure of mental energy in the mere act of listening to verbal articulations, or in that silent repetition of those which goes on in reading - the perceptive faculties must be in active exercise to identify to identify every syllable", etc.

But in examining the laws of style Spencer necessarily speaks of the hearer (recipient) only and says nothing about the speaker (producer). Now I found that in valuation of a language, or a linguistic expression, both sides should be taken into consideration: the best is what with a minimum of effort on the part of the speaker produces a maximum of effect in the hearer. This is the substance of my essay "Fremskridt i sproget" (1891), which formed an introduction to my thesis "Studier over engelske kasus", and was expanded in English in "Progress in Language" (1894) and still more so in "Language" (1922).

When some years afer the first appearance of my theory W. Ostwald began the publication of his philosophy of energetics, I recognized in his ideas the same point of view that I had already applied to language: I found in this coincidence a strong argument in favour of my views (see "Energetik der Sprache" (1914), reprinted in "Linguistica").

"Survival of the fittest" - this is the ingenious watchword invented by Herbert Spencer to explain what Darwin understood by "natural selection": those individuals of a species are preserved that are best adapted for their environments. Can this be applied to language? Evidently not to languages as wholes: which of these are preserved and which are doomed to extinction is determined by other considerations than the intrinsic perfection of their structure or the reverse: here wars and political conditions are generally decisive. But within a language we must admit the truth of the slogan: those particular traits of a language which are best adapted to their purpose tend to be preserved at the cost of others which do not answer the linguistic purpose so well. This will be demonstrated in many particulars of the following disquisition.

1.2 When I began writing on language, the prevalent theory was this: language had begun with inflexible roots, some of these in course of time became subordinate grammatical implements which were first agglutinated to and eventually fused with the more substantial elements. In this way was achieved the development of inflexional . languages such as primitive Aryan (Indo-European, exemplified in Sanskrit, Greek and Latin); here the high-water mark was attained, and since then we witness only decay, degeneracy, and destruction of the beautiful structures of these old languages. To this I objected, trying to show that viewed from the point of view of human energetics so far from being retrogressive the tendency in historical times has on the whole been a progressive one.

Though it is possible that in my endeavour to refute old theories I paid too little attention to those changes that are not beneficial, I never maintained that all linguistic changes in all languages and at all times made for progress; I never was an "optimist à la Pangloss", but I still think that I was right in saying that on the whole the average development was progressive and that mankind has benefited by this evolution. (See the detailed exposition in Lang., p. 319-366.)

In the summary found ib. p. 364, I said that the superiority of the modern Aryan languages as compared with the older stages manifests itself in the following points:

(1) The forms are generally shorter, thus involving less muscular exertion and requiring less time for their enunciation.
(2) There are not so many of them to burden the memory.
(3) Their formation is much more regular.
(4) Their syntactic use also presents fewer irregularities.
(5) Their more analytic and abstract character facilitates expression by rendering possible a great many combinations and constructions which were formerly impossible and unidiomatic.
(6) The clumsy repititions known under the name of concord have become superfluous.
(7) A clear and unambiguous understanding is secured through a regular word-order.

Each of these points had in the preceding pages been fully documented by typical examples; no. (2) , for instance, through reference to the chapter in "Progress" in which the case system of OE and ModE had been tabulated in the same way, filling seven and two pages respectively. With regard to (3) I pointed out the very important consideration that when we look at the actual facts we see that anomaly and flexion go invariably together (Lang. 232): it is thus wrong to say that "the Aryan inflexions were once more numerous and at the same time more distinct and regular," as Sweet says (Collected Papers 68).

These chapters in my book have never been refuted, either as a whole or in detail. Most subsequent writers on language simply disregard the question of progress or retrogression, or even mention it as lying outside the sphere of scientific linguistics.

In reading many books on the history of language one gets the impression that the history of languages is nothing but a purposeless fluttering hither and thither. I tried, and shall again in this treatise try to show that a great many changes manifest a purpose, conscious or unconscious, to better existing conditions, and that some changes, though apparently detrimental, may, if summed up, in the long run prove beneficial and make for progress. People have sometimes blundered into improving their mother-tongue.

1.3. The only two writers, as far as I know, who after me have dealt at some length with the question of progress in language are Charles Bally and J. Vendryes. The former discusses it in Le Langage et la Vie, 1st ed. 1913 (thus nineteen years after Progress), 2nd ed. 1926 (thus three years after Language). He has no difficulty in showing that language has not attained, and on account of the multiplicity of practical life probably never will attain, the complete logical idea of univocité - the same idea always expressed by the same form, and the same form always meaning the same thing - and he goes on to examine the relation between synthesis and analysis with examples of their mutually replacing each other so that advance is a pure illusion: it is six of one and half a dozen of the other; the whole linguistic development is made up of rhythmic ups and downs. He mentions neither my previous work nor my criticism of his 1st edition. His treatment is unsatisfactory because he does not compare the structure of earlier and later stages of the same language as wholes, as I had done in the chapter mentioned above (1.2), and because he does not see the importance of the point of view of energetics, the relation between the output of energy and the result attained.

J. Vendryes in the last chapter of Le Langage (1921) deals with "le progrès du langage". But though he warns against "une confusion fâcheuse entre la langue littéraire et la langue tout court" he does not seem himself to have avoided this confusion. His main result is that on the whole gains and losses counterbalance one another very nearly: everything depends on the hand that shakes the instrument. He no more than Bally has seen the importance of energetics, nor does he compare two stages of one and the same language as wholes. Most of his chapter does not concern us in this connexion.

1.4. In a very short chapter of his admirable book The Making of English Henry Bradley speaks of "Profit and Loss". He turns against some extreme optimists who think that in the evolution of language "everything happens for the best, and that English in particular has lost nothing, at least so far as its grammar is concerned, that would have been worth keeping". But who are these optimists? As already remarked, I myself never said that everything happened for the best. Bradley says that in writing English special care and ingenuity are often required to avoid falling into ambiguities - but is not that true of other languages as well? In colloquial English there are some abbreviations which sometimes occasion inconvenience by their doubtful meaning: thus he's may be either 'he is' or 'he has', and I'd may be either 'I had' or 'I would' - but certainly in nearly every case the form of the following verb will resolve any doubt. Still, Bradley says that English has gained by many additions to its grammatical resources and by the disappearance of superfluous inflexions as well as by the reduction of those which remain to mere consonantal suffixes; this has greatly increased the capacity for vigorous condensation. English thus has the peculiar advantage of a noiseless grammatical machinery, and further the ability of stressing the auxiliary as in 'I did live there' and of using the auxiliary by itself as in 'Yes, I do,' 'it certainly will not.' I think that in spite of his cautious expressions his few pages on the subject justify me in enlisting the late scholar among those who on essential points agree with my views on progress in English.

G. Cederschiöld's paper Framsteg i språket (1897) reprinted in Om kvinnospräk och andra ämnen (1900), fully agrees with my points of view. So does E. H. Sturtevant in Linguistic Change (1917), p. 176.

1.5. In this treatise I shall not repeat the substance of what I said in Progress and Language, but merely in detail examine some points in which the progressive tendency manifests itself in various ways. In thus taking up some related strands and trying to weave them into a new pattern I am afraid that readers of my other books will here and there recognize ideas and examples they have seen elsewhere, but as they are given here in a new setting and for a different purpose I hope I may be forgiven for such unavoidable repetitions.

Back to Efficiency in Linguistic Change contents
Back to International Auxiliary Languages

James Chandler 2001