1150 ^62 UNIVERSITY 1160 m-. fyxmW mm BOUGHT % F SAGE EN DC TH Hcnr HOME USE RULE«. All Books subject to Recall. Books not needed for instruction or re- search are returnable i within 4 ^eeks. Volumes of periodi- cals and of pamphlets are held in the library as much as possible. For special purposes they are giveti out for a limited time. Borrowers should not use their library privileges for the bene- fit of other persons. Books not needed during recess periods should be returned to the library, or arrange- ments made for their return during borrow- er'sabsence, if wanted. Books needed by more than one person are held on the reserve list. Books of special valu6 and gift books, when the gxver wishes -it, are not allovyed to circulate. Marking books strictly for- bidden. Readers are asked to report all cases' of books marked or muti- lated. yiZ.ZO^VS'ao 7-7 ^i-Z- -y^.g.<...... 7673-2 THE ^he Problem of Spelling Reform By The Rev. Professor W. W. Skeat Fellow of the Academy Cornell University Library PE 1150.S62 3 1924 026 632 202 ( Frovi the Proceedings of the British Academy, Vol. II) New York Oxford University Press American Branch 91-93 Fifth Avenue. London : Henry Frowde. Copyright, 1906, by lb University Press American Branc IK- ^o^cg-^o THE PROBLEM OF SPELLING REFORM By the rev. professor W. W. SKEAT FELrOW OF THE AcADEMY Read May 2, 1906 I SUPPOSE I may assume that the British Academy will at all times be ready to consider questions that relate to general education ; and I am personally of opinion that there are few subjects of general interest and importance that are more deserving of attention than the often debated one of Phonetic Spelling. I do not propose to waste the time of those who are so kind as to listen to me, by stating or recommending any views that I have formed for myself, during the many years in which this subject has never been far from my thoughts; my object is rather to review the present state of the question, and to consider what progress, if any, has been made in the direction of its solution. Notwithstanding aU that has been said and written upon the subject during the past forty years, I fear it must be admitted that we are still in very much the same position as we were at the begin- ning of the discussion ; at any rate, as far as relates to the attain- ment of any practical results in this country. But as regards the theory of it, something has been achieved. The labors of Dr. Ellis and Dr. Sweet have by no means been lost ; and, in order to see what advances have been made, a slight sketch of the history of the move- ment may prove useful. When it was first proposed to improve our modem system of spelling, it soon dawned upon all serious enquirers that the first step was to examine the meaning of the symbols which we employ for the purpose of representing the spoken language. There was a diffi- culty at the very outset ; for the first impression which the very look of the written language is apt to produce — the impression, indeed, which it actually produces upon the minds of a vast multitude amongst us — ^is that the whole matter is inexplicable, and is not likely to be capable of explanation. I remember very well, for example, meeting with a pas^ge in the works of so popular an author as Bulwer Lytton (I have unfortunately lost the reference) or 2 PROCEEDINGS OF THE BRITISH ACADEMY in which modem English spelhng is frankly declared to have been the invention of the devil; because it is unlikely that any other being could have hit upon a system which, to all appearances, is so chaotic, so inconsistent, and so contradictory in many instances to all common sense. Of course this opinion was not meant to be taken seriously; but it fairly represents a very general conviction, viz. that the whole system is so capricious as to be likely to defy any general comprehension of it. And it is only right to admit that, before we can undertake to correct our spelling with any prospect of sufficient success, we must first of all endeavor to understand and to explain it as it is. The right objection to make against Bulwer Lytton's sweeping allegation is to say that it is unscientific. For, after all, spelling is a merely human contrivance, and must be capable of explanation by historical methods. It may be a laborious business to explain the modern spelling of every English word, but of course it can be done ; and perhaps it is hardly necessary to add that, as a matter of fact, the history of every verbal form is being regularly recorded in the New English Dictionary; in which it is proved, beyond all doubt, that the whole problem admits of a complete and exact solu- tion, and is such as can, in most instances, be easily understood by experts. In order to be an expert in this matter, it is necessary to study not merely the history of the written forms, but also that of the sounds which such forms were intended to represent. This is- at the bottom of the whole question, and until it was apprehended that such is really the case, no real progress was possible, and therefore none was made. All that could be done was to note a few unwar- rantable caprices which a very little experience could easily correct. I take as an easy case the curious craze that was fashionable in the seventeenth century of writing sc for s before i, or sometimes before e, at the beginning of a word. This was of course due to the fact that there is a silent c in the initial sc of the French word science; so that it became 'scientific' to extend this system of spelling words to other cases. John Ray, in his Note on the Errors of our Alphabet, makes particular mention of this in 1691. He says: 'I might also note many false spellings in particular words, as scitvMte for situate, which is but lately come up, and hath no ap- pearance of reason, the Latine word being situs, without any c. Scent for sent, signifying a smell or savour, which writing is also but lately introduced, and hath no more ground than the former the Latin word from whence it comes being sentio.' We are now happily rid of the form scituate,hut scent is still with THE PROBLEM OF SPELLING REFORM 3 us. Not only so, but modern English still clings, senselessly, to several other forms that are equally ridiculous and indefensible. Take for example the word scythe, which appears in Shakespeare's Plays as sythe (Hen. V. v. 2. 50 ; L. L. L. i. 1. 6 ; Ant. iii. 13. 194). and in his Sonnets as sietJi or syeth (Son. 12, -60, 100, 123). It is true that Richardson quotes the modern spelling scythe from Shakespeare's Lover's Complaint; but the quotation lies open to two objections. One is, that many of us doubt whether Shakespeare wrote that particular poem; and the other is, that a reference to the beautiful facsimile edition lately published by the Oxford Press shows that Richardson was mistaken ; the form which appears there is sithed (1. 12), which is meant for a past participle, with the sense of 'mown.'^ Another monstrosity is scion, a spelling which also appears in modern French, a language whose spelling, like our own, abounds with inconsistencies. But Cotgrave's French Dictionary gives the alternative spelling sion, and the Old French had the form cion. Whether it is really allied to the French scier, to saw, to cut, is now much doubted, but even that would not help it out, for there is no c in the Old French sier, which is derived from the Latin secare, to cut, with a total disappearance of the guttural consonant, j ust as in the French nier from the Latin negare. And finally I might instance the extraordinary form scissors, which simulates a derivation from the Latin scvndere; though the old spelling cisours connects it with the French ciseau and the English chisel; from the Latin caedere, to cut. But the chief point which we must always bear in mind is really this ; that the mere observation of old forms will not carry us far, unless we can succeed at the same time in realizing how those old forms were pronounced. This is the one fruitful idea which has taught us so much, has established modern etymology on so firm a basis,^ and has led to so great an advance in the philology of var- ious languages. It is the one lesson which is essential, and is never- theless realized by very few of us. We all need to recall to our minds, from time to time, the elementary fact that all languages are essentially spoken languages ; and that the written words are mere symbols, employed to represent real spoken sounds, and some- times performing their office in a very slovenly and unsatisfactory manner. For indeed, the spoken forms change slowly from one generation to another, yet so imperceptibly that the necessary accompanying changes in the written forms are not kept up to date, but frequently remain unchanged for centuries, until such propriety 1 ' Time had not sithed all that youth begun." 4. PROCEEDINGS OF THE BRITISH ACADEMY as they once possessed has gradually faded out of them. The most extraordinary example of this truth is obvious in the case of Latin. The old pronunciation of Latin is sufficiently well known to scholars, but is now very lightly regarded, and nowhere more lightly than in England. The Italians have, for the most part, preserved the vocalic sounds and many of the consonants, by the nature of the case ; but I fear that many of them pronounce Latin as if it were mere modern Italian, and take no further pains about it. The Germans have a tendency to pronounce it as German, but probably do more justice to the right sounds than others do. Spaniards are apt to confuse it with Spanish, and Frenchmen with French. But however this may be, what most concerns us is the treatment of Latin in England. It seems to have escaped the observation of many that there was a time in England when Latin pronunciation was not far wrong. To this day the Latin c is pronounced like k in Wales, even before e and I; as in such common words as ceffyl, a horse, allied to caballtts, and ci, a dog, allied to ccmis. The same sound was so universal in England before the Conquest that the letter k was but seldom used, as being hardly wanted. The Roman Cantium has become the English Kent; but it is a most interesting fact that the most universal spelling of Kent in Anglo-Saxon MSS. is Cent, with a C. The Anglo-Saxon vowels were denoted by Latin symbols, precisely because they were sounded like the Roman vowels, as in the famous word win (pronounced as E. ween) from the Latin uinum, which even preserves the right initial consonant to this day, though the modem form is wme. In Chaucer's time, Latin was still pronounced as English, and yet pronounced reason- ably well, because the vowel-sounds were then but little changed. But as from time to time the sounds of English altered, Latin was still sounded as English, and suffered precisely the same changes,; and this is really the reason why many people believe that Latin is quite properly pronounced when it is pronounced like the English of the present day. They forget how much has happened since the days of Alfred. They still sky Kent for the form which we once wrote Cent; but they no longer say kentum for the form which we stiU write centum. I suspect this to be at the bottom of the whole trouble. My memory goes back to the time when it was a matter of common belief that the sounds which we employ in speaking English have never altered! Indeed, this was quite a logical conclusion; for if it be correct to pronounce Latin like modem English, it must follow that we must also pronounce Anglo-Saxon after the same fashion, since it employed the same alphabet as the Latin and apparently THE PROBLEM OF SPELLING REFORM 5 used the same symbols in the same way. In such a belief I was accordingly brought up, and in this benighted state I remained until Dr. Ellis, in 1869, brought out the first part of his great work on Early English Pronunciation, with especial reference to Shake- speare and Chaucer. He began by showing that the question of our pronunciation was actually discussed by many writers of the six- teenth, seventeenth, and eighteenth centuries, and that it was impossible to reconcile their statements with the present state of things. And when- once the old belief as to the stability of our vowel-sounds had thus been shattered, the whole position was changed ; and all that remained — though this was a large remainder — was to ascertain what was the nature of the sound-changes, and to fix convenient dates at which the sounds employed could be most usefully investigated. The chief result of Dr. Ellis's work was to show that the question of Elizabethan pronunciation offers great difficulties ; but that, as we recede from the present date, such diffi- culties readily decrease. The question as to the Chaucerian pro- nimciation is much simpler; and we are very much helped in this matter by the extremely careful and almost rigid exactitude which was observed both by Gower and Chaucer in their methods of riming. The older MSS. of the thirteenth century are even clearer, as a rule, in their mode of denoting vowel-sounds; though we also come across MSS. which were evidently the work of some Norman scribe who experienced much difficulty in struggling with sounds so strange to him as the English th and sh and wh. As for the wh, the Normans successfully persuaded the inhabitants of Southern England to exchange it for w; and at the same time to neglect the guttural sound which they were willing to represent by gh, though they declined to pronounce it themselves. When we get back to still earlier times, viz. to the period before the Norman Conquest, the problem is still further simplified; we find that, as regards the vowel-sounds, we are dealing with symbols such as are still in use in Italy, and that we have, in fact, got back to the time when Latin was pronounced as English, and at the same time pronounced cor- rectly. If we turn to p. 634! of Dr. -Ellis's work, we find that his con- clusions as to Anglo-Saxon pronunciation do not differ materially from those which had already been put forward by Rask, the Dan- ish scholar, as far back as 1817, though very little attention was paid by us to Rask's teaching before 1830, when an English trans- lation of his Grammar, originally written in Danish, was printed at Copenhagen, the translation being the work of Benjamin Thorpe. Perhaps the earliest book printed in London in which Rask's results 6 PROCEEDINGS OF THE BRITISH ACADEMY were made generally accessible was Vernon's Anglo-Saxon Guide, printed by J. Russell Smith in 1850 ; a book which remains a good one to the present day, though it has been superseded for general use by Dr. Sweet's well-known Anglo-Saxon Primer. Later than this monumental work by Dr. Ellis, we have the important History of English Sounds from the earliest period, with full word-lists, published by Dr. Sweet in 1888. In this the order of investigation is very properly reversed. Dr. Ellis traces the history of our sounds backwards, from the sounds of modem English to those of Anglo-Saxon. Dr. Sweet was thus enabled to exhibit the development of our sounds in tHeir historical order, beginning with Anglo-Saxon times and tracing the changes introduced during the various periods which he calls the Middle English period and the First, Second and Third Mod- em periods; the last of which brings us down to the nineteenth century. Two other books by Dr. Sweet also deserve mention in the present connexion. These are his First Middle English Primer, which discusses the pronunciation of the Ormulum and the Ancren Riwle near the beginning of the thirteenth century ; and his Second Middle English Primer, which discusses the pronunciation of Chaucer. Besides these works, we are fortunate in possessing an excellent treatise on the development of what has been not very happily termed Lowland Scotch, though it would have been better to have termed it Northumbrian English. I refer to The Dialect of the Southern Counties of Scotland, by Dr. Murray ; published by the - Philological Society in 1870-72. It will be observed that he is careful to avoid the more usual name. I have given these details because they are essential to the history of spelling reform. The intelligent examination of the older forms of words soon showed that not merely their sounds, but the spellings which denoted those sounds, have changed from time to time; and further — which is much to the point — that many changes made in spelling were made from a phonetic point of view, the intention being, in such cases, to get a better representation of the sounds that were intended to be suggested. In days when there was no fixed standard, many such attempts were unintelligent and unworthy of serious regard; but we also come across manuscripts, such as the celebrated autograph copy of the Ormulum and the Ellesmere MS. of Chaucer, in which symbols are employed with great regularity, so that much can be learnt from them. Dr. Sweet's History of English Sounds contains two valuable word-lists. The first of these contains four columns. 'Of these, the first shows the A.-S. THE PROBLEM OF SPELLING REFORM 7 written form; the second, the Middle English written form; the third, the modem English written form ; and the fourth, the mod- ern English spoken form,- as expressed by a special phonetic nota- tion invented for the purpose. For example, the word great gives us (1) A.-S. great; (2) Middle English greet or greet; (3) mod. E. great; (4) the modern sound, expressed in phonetic spelling as greit. The fact is, accordingly, that so far from the idea of pho- netic spelling being modern, it is extremely old. For it is obvious that those who first undertook to reduce sounds to writing could not possibly have taken as their guide any other standard than the sound of the spoken utterance. The notion of unphonetic spelling, i. e. of spelling a word one way and pronouncing it another, has grown upon us unawares, merely because we have been too lazy or too in- different to reform our spelling from time to time with sufficient frequency. The most probable reason for this is the unpleasant fact that many of the most violent or most sudden changes in the spoken language have taken place in comparatively recent times, when the appearance of the printed forms had become familiar to many readers ; and the required alterations would have been resisted by the printers and disliked by many who, having learnt to read already, had no particular regard to the convenience or edification of a coming generation. The very word great, already referred to, affords an excellent illustration. The A.-S. form great was purely phonetic. The gr was followed by a diphthong expressed by ea; i. e. the former element was a long e, probably an open e resembling the modem English a in Mary, immediately followed by a short a like the Italian short a in the first and last syllables of amata. But in Early Middle English the diphthong became a vowel, by the grad- ual diminution and neglect of the latter element, so that we find in the Ormvlum the spelling grat, with two dashes or slanting strokes above the ee, to call attention to its length. The use of the ce was to denote that the e was open, the close e being written as e. Then came the most important spelling reform in the whole history of our language, when it was completely respelt by Norman scribes upon French models, chiefly during the reigns of Edward I and Edward II. These scribes were unused to the A.-S. symbol ae, and took the unwise step of abolishing it ; with the result that they had but one symbol, viz. ee, to denote the long sound of e, which was used, with- out distmction, for both the close and open varieties of the vowel- sound. The consequence was that the spelling of the word was Normanized into the form greet, as in the EUesmere MS. of Chau- cer. If this adjective was used in the plural, it was common to 8 PROCEEDINGS OF THE BRITISH ACADEMY write grete instead of greete, as it was supposed that the length of the vowel was thus sufficiently indicated. But at the very same date we meet with a duplicate form grete in -which the e was close, the sense of which was 'I greet' ; and this ambiguous state of things became at last intolerable. For indeed both sounds were unstable, and subject to continuous slight changes. Of all our long vowels, the long close e seems to have been the first to approach its modem English value, with the natural result of creating confusion throughout the whole scheme, and a general diminution in the pho- netic values of the symbols. Already, about the year 1500, when we enter upon what is now generally called the modern period, as distinguished from late Middle English, it would appear that the words great and greet, though still spelt alike, viz. 'greet,' had approached more nearly to their modem values, with a very distinct and easily apprehended difference in their sounds. About this time a new spelling reform set in, viz. ( 1 ) to restrict the use of ee to the cases in which the e was originally close and was now ap- proaching the sound of the modern E. i in machine, and (2) to use the symbol ea for the sound developed from the open e. Already in Palsgrave (1530) and Tyndale we find great as a usual spelling; and from this time forwards we find the sounds resulting from the two varieties of the M. E. ^ fairly well distinguished in writing. In the time of Pope, it is well known that we find tea to represent what we should now spell tay; and a similar pronunciation is still heard in Ireland. But about the year 1800 the pronunciation of tea as tay became archaic, and the sounds which had been distin- guished for so many centuries at last began to approach one another, and at the present day (except in a few instances) are precisely alike. The chief exceptions are two, viz. great and break; for although there was at one time a tendency to change these along with all the rest, and a few people pronounced these words as if they were spelt greet and breek, these sounds did not prevail, and are now very rarely heard except in the dialects of about fourteen counties. The reason for this curious anomaly is doubtless, as Dr. Sweet points out, that in both cases the diphthong (phonetically denoted by ei^) is preceded by an r that is further strengthened by a preceding consonant. For it is a fact that the sound of r frequently affects a neighboring vowel. From the above account it naturally follows that the present habit of pretending to make a difi'erence between 'ea and ee is an obsolete absurdity. We may, indeed, usually conclude frdm the modem spelling that such words as deal,' and clean, and heat, were ^ i. e. the ei in ' vein ' THE PROBLEM OF SPELLING REFORM 9 pronounced in the days of James II as if they were now spelt dale and clane and hate, i. e. as they are still pronounced by many in Ireland ; but this has nothing to do with modern literary usage, and we should all be gainers if we abolished a now useless distinction, and spelt these words with double e, in conformity with the rest. Such spellings as deel, cleen, and heet would be perfectly intelligible at the first glance to any child who can read at all ; whilst we might retain break and great for the present. I here purposely take occasion to consider an objection which is frequently raised by such as do not understand the true object of spelling. The moment that we propose to write cleen for clean, in accordance with common usage, up jumps the objector who has been lying in wait for this very opportunity, and says : — 'Then you would actually propose to represent the mighty ocean by such a spelling as see! How then shall we be able to distinguish it from the verb to see?' The triumph with which this supposed 'poser' is put is almost piteous to witness ; for it shows but a very limited ac- quaintance with the English language. The answer is: — 'In the same way as you distinguish ball, a dance, from ball, a sphere ; or be- tween a bark that is a ship, and the bark of a dog; or between the hundreds of other homonyms that fill twenty pages, in double column, of my Etymological English Dictionary. It is the spoken language, once more, that alone matters, and there is always a con- text. If two such words as sea and see require to have different spellings, how is it that they do not require to have different pro- nunciations.'" It is the old mistake of supposing that an object seen imperfectly in a mirror is more original and valuable than the object itself. In consequence of the results as to the history of English sounds that had thus been obtained by Dr. EUis and Dr. Sweet, it was no wonder that it occurred to the Philological Society, with whom they were so closely connected, to consider the whole question of Spelling Reform afresh; and it did not take very long before the following results were obtained. 1. Thri best scheme of all would be one in which the symbols would be referred to their original values; or, in other words, in which the vowel-symbols, in particular, should be used as they are now used in Italy and were used of old by the Romans from whom we borrowed our alphabet at a very early date. At the same time, it would be advisable to adopt new symbols for such simple sounds as th and sh, which are now somewhat clumsily denoted by combina- tions. That such a scheme is eminently practicable is fully demon- strated by its use by Dr. Wright in compiling his English Dialect 10 PROCEEDINGS OF THE BRITISH ACADEMY Grammar. For it is obvious that a scheme which is capable of representing all the pronunciations found in all our English dia- lects must be capable, inclusively, of representing all the sounds heard in the literary language only. 9.. It may be admitted that there are two practical objections to this course ; one is that it requires a certain amount of new type ; and the other that the English people, speaking generally, are so hopelessly ignorant of phonetic principles that they are incapable of understanding symbols used with their true value. Even those who admit that they pronounce the i in machine ,Vik& the ee in meet are usually incapable of applying such a principle generally; so that if one were to denote the word meet by the form mlt (or miit) they would still argue that this last form must be pronounced as mite. S. We are thus led to consider the question anew. Suppose we retain, as far as they are suitable, the present symbols, can anything be done to reduce the present state of chaos to something that more nearly resembles a state of decent order.'' This question was considered by Dr. Sweet in all seriousness. A paper which he wrote with the title of 'Partial Corrections of English Spelling aproovd of by the Philological Society, 1881' appeared in the Supplement to the Transactions of the Philological Society for 1880-1 ; and, though it occupies but thirty-eight pages, it is extremely clear and full, and contains a list of a large number of representative words which undoubtedly admit of very considerable improvement. This is, in fact, an initiative instal- ment of Spelling Reform such as we should be very fortunate to obtain ; for it is a great improvement upon our present usage, and is carried out on perfectly consistent and sane principles. By way of example, I here note a few simple cases in which our spelling might be improved, all of which are duly noted by Dr. Sweet. 1. The first is the abolition of the silent final e in cases where it is useless. There is an extremely silly habit of never allowing v to stand at the end of a word. Such is our conservatism that the origin of the habit dates from the time when the sound of v was denoted by u, and when the u was always followed by an e, or at any rate by a vowel, in order to show that v ( and not u) was the sound intended. When the old form haii£ became have, the final e per- formed no function, and should have been dropped. We ought accordingly to have no hesitation in writing hav, giv, dbwv, cvm, and the like. We could then distinguish liv from live (with long i). We could further extend this dropping of e to words like THE PROBLEM OF SPELLING REFORM 11 agreev, aproov, solv, adz, frees, and many others. Awe may be re- duced to aw, and axe to ax. It is interesting to know that the final e in ax{e) isunjustifiableby etymology, as is duly pointed out in the New English Dictionary. To write axe is due to mere caprice ; especially when we write ox instead of the Mid. Eng. oxe, in which the final e was really syllabic and so had once a meaning. 2. In a large number of words ending in -le, the final e is quite needless ; we can easily tell how to pronounce such forms as assembli, litl, dvbl, heetl, icicl, promis, depnit, activ, handsum, therefor. It is also better to drop the e, even when not final, in words like drivn, writn; and similarly to drop the o in butn. 3. The use of ea to denote the sound of short e is absurd and troublesome. We should certainly write bredth, medow, brekfast, hed; with a great many more. For heart we should write hart. The use of eo for e in jeopardy and leopard should certainly be given up. The word 'people' is represented in Dr. Sweet's scheme by peple. An alternative form is peepl, which would be more con- sistent with other amended forms. 4. The use of ie for ee is unhistorical, and should be discontinued. Examples are: — acheev, beleev, cheef, feeld, seege. Sieve, like sift, was once spelt with simple i; so that siv is absolutely correct. I may here observe that Dr. Sweet omits to notice the ei in receive; yet this is equally useless ; we really say receev, deceev. 5. The Tudor form oo should be restored in words like improov, looz, moov. 6. The original Old Norman u in such a word as cumfort was turned into o by the later Norman scribes merely because lum was indistinct in manuscripts ; on which account they wrote comfort for the greater distinctness ; i. e. they used o for short u before m and n and u, as in comfort, monk, money, cover (formerly cotter). But modem printing is clear enough to admit of the restoration of original forms; such as cumfort, cwmpany, cwm, mwnk, mwney, cwver. 7. In many words ou is' written where the modern short u is meant. We should therefore write burgeon, cur age, cu-zim, flurish, jurney, rwf, tuch. So also in words ending in our, it suiBces to write or; as in labor, honor, harbor. It is difiicult to understand the position of those who defend such a spelling as labour. They pretend that it is etymological, as it certainly was once, when the word was accented on the latter syllable ; but it ceased to be so about 400 years ago, when the accent was thrown back upon the former syllable. They also conveniently forget that the change from final -our to -or has already been la PROCEEDINGS OF THE BRITISH ACADEMY silently made in several words, such as emperour, horrour, terrour, errowr, mirrour, jurour, confessour, cis sours (now spelt scissors), and others. The objectors would be equally shocked to see the old spelling restored in these words, which shows that they are incon- sistent and insincere. Another favorite argument, quite as untrue as it is unpolite, is to call the form labor 'American' ; for the spell- ing laborer (for labourer), in six MSS. out of seven, in Chaucer's Knightes Tale (A 1409) clearly shows that our ancestors knew how to represent the -our- of an unaccented syllable before Columbus was bom. 8. The final ue in decalog(ue), demagog(ue), and the like is useless and modem; but may be retained in fatigue and vague, where a' strong stress immediately precedes it. 9. The doubling of consonants is often unnecessary. Such forms as ad, eg, od, bailif, buz, ful, stif, are quite intelligible. So also in bail, ketl, wrigl, writn, traveler. In many words, the double letter is unhistorical and unetymo- logjcal ; as, e. g., in arrive, from the Old French ariver. We should do better to write arive, adres, aju^t, aquit, of air, atak, comand. Few people who look at the form affair would guess that it simply resulted from the French phrase a faire. The double m in com- mence, from 0. F. comencer (Ital, comi/nciare), is ignorant and ludicrous. 10. It was by an unintelligent trick, at the time of the Renais- sance, that an unoriginal b was thrust into debt both in French and English. The French have now got rid of it, but John Bull clings firmly to his debt. It is simpler, as well as more correct to write det, detter. The final b is useless in lamb, limb, numb, thvMib. So a,Jso plu/mmer is better than plumber^ and is so spelt when used as a surnaine. 11. Some words have gone quite wrong; we actually use ache for alee; and the old spelling of anchor was anker. As to choir and quire, the latter form alone denotes the sound, and should be pre- ferred. 12. An e is often needless when a word ends in -ed; as in pulled, which might be reduced to puld. There is no more objection to be raised against lookt or slipt than there is against spUt or left. I have not noticed all the points which Dr. Sweet discusses, but the above are sufficient to show how much might be done to make our spelling more phonetic. The usual outcry as to the loss which such improvements would cause to the cause of etymology is, for the most part, as false as it is insincere. It is only raised for the sake of hindrance, and every etymologist who has investigated the THE PROBLEM OE SPELLING REFORM 13 question fairly and without prejudice, and is sufficiently familiar with the history of the older forms of the language, knows per- fectly well that many of the above simplifications actually restore older forms, and are often much to be commended by the ety- mologist. At this point I pause awhUe, in order to point out two great advantages that would certainly result from the introduction of even such small instalments of spelling reform as those to which I have just drawn attention. The first is that these partial reforms would necessarily involve the disuse of a large number of useless letters. In this way more matter would be got into a page, and some labour in the composi- tion of the type would be saved ; and as this would happen in every case, whether the book be small or great, it might very easily save every printer and publisher a considerable sum of money. It would not be surprising if the aggregate saving, in the course of a year, throughout the British Empire, were to amount to some thousands of pounds. Every one would gain by it ; the writer would have fewer letters to write, the printer would have fewer to print, and the reader would have fewer to spell and read. Many ambiguities would disappear, and the written record would exhibit a closer approximation to the sounds which it professes to represent. The second is that the task of learning to read would be consider- ably simplified, and the time taken to achieve that task would be considerably shortened, to the equal delight and profit both of teacher and pupil. In this case there can be no doubt at all that the sums thus saved to the nation would be very considerable. I now come to the consideration of the present state of the prob- lem. The above-mentioned suggestions for partial reforms in spelling, which were' intended merely to remove such uses as are wholly obsolete or unmeaning, were put forward by the Philolog- ical Society in 1881, just a quarter of a century ago, at a time when there was a very general desire on the part of intelligent peo- ple — particularly with regard to the great drudgery and expense which the modern system of teaching children to read and write in- volves — to lighten a very great and almost intolerable burden, and to save time for instructirin in such elementary subjects as are useful and necessary even in the case of the poorest. Considering the great interest which many take, or profess to take, in the highly im- portant matter of elementary education, it will readily be under- stood that any "proposal to simplify such education, and in particular a proposal such as this — whereby it might very 14 PROCEEDINGS OF THE BRITISH ACADEMY reasonably be expected that the time allotted to learn- ing to read would be reduced by at least one fourth, and probably even by half — would receive encouragement, or at least a patient hearing, by such as had gone through the drudgery them- selves and had a true feeling of compassion for their pooret and more helpless brothers. But the result, I speak it to our utter shame, has been precisely the contrary. There was not a news- paper, journal, or periodical in the whole country (with a few un- influential exceptions) that was not resolutely determined to crush any such attempt ; and the same resolution exists at the present day. Any other proposed reform could have obtained, and could still obtain, some sort of recognition, but not this one. The merest hint that it would be wiser to spell Uv without a final e, in order to distinguish it from the adjective live, was enough, and still remains enough, to rouse a storm of indignant denunciation. Of course there is some reason for this, and it is not difficult to discover it. It simply amounts to this. The indignant writers were discuss- ing a subject which they had never studied, and which they did not unders^tand ; and they did so with perfect honesty, because they were not in the least aware of their own ignorance. Such is not the case with such other subjects as are most frequently discussed. The man who does not know botany at least knows that botany is a science that requires to be learnt; and consequently, since he has not learnt it, he must refrain from writing upon it, because some readers will only laugh at him, and the rest will disregard what he says. The same is true, not only of every scientific study, but of most literary subjects as well; the chief exceptions being etymology and phonetics. As for etymology, it was once notoriously the custom to regard as an oracle the wildest guess of the boldest and most ignorant guesser; but most of our respectable journals have exercised such fairness as to pfermit such ignorance to be exposed; so that the worst offenders have, for the most part, learnt some caution, and have no longer any certain assurance that they will cover themselves with glory by the simple process of guessing at haphazard. But with phonetics it is far otherwise. They can only be truly taught or explained by the Hving voice,; and they can only be under- stood, upon paper, by such as have received rather careful instruc- tion. And without some knowledge of phonetics, our spelling can- not be understood or intelligently discussed. And further this simple fact, though it is obvious to one who knows something of the matter, is wholly unknown to the mass of even the most highlv educated of mankind. It may be doubted whether one in ten THE PROBLEM OF SPELLING REFORM 15 thousand scholars can answer such a simple question as this: — . 'How came, the symbol ou to represent the sound of the diphthong which is now heard in the word house, or the symbol ow to represent the same sound in the word cowf Or again — 'How is it that the symbol oo, which obviously ought to represent a double or long o, is actually employed to represent a double or a long u, as e. g. in the word cool,- which now rimes with rule?' Yet the very men who are wholly unable to answer such questions as these, questions that were once unanswerable but are now familiar to all who know the elements of phonetics, are the very men who presume to sit in judgment on the forms which they do not undei^stand. Still worse, they will not endure to listen to any instruction or to any argument, but dismiss every attempt to represent a spoken word by a more rational sym- bol with the sole argument which appeals to them, that it 'looks so ugly.' This is, in fact, what it has come to. The Englishman who has learnt, in the true Chinese fashion, to associate every spoken sound with a fixed and unalterable symbol, however inappropriate or un- intelligent it may be, no longer uses his ear, but depends solely upon his eye. For him the sole question is, not how does the word sound, but how does it look. He learns his symbols by heart and by rote, with regard only to their visible appearance. His task is less stupendous than that of the Chinese, because, after all, a large nimiber of our words are phonetically spelt ; yet it is remarkable that it never occurs to him that there is no reason whatever why they should not all have the same merit. The more senseless and uncouth the spelling, the greater is the effort to learn it ; and it is precisely because it 'looked so ugly' to him when he first mastered it that it looks so beautiful now. This very charge of ugliness is an unconscious confession that the critic has ended by worshipping what he once abhorred ; and, rightly considered, is a severe condemnation of the whole system. If, instead of thus yielding to prejudice, we should dare to use our common sense, it would be obvious that the best spelling of a given word is such a combination of symbols as wiU at once suggest the sound of it. It is possible to pursue the consideration of this matter yet a step further. Whence, we may ask, does this general ignorance of phonetic principles, on the part of Englishmen, arise.'' I have no doubt that the right answer is one which may not readily suggest itself, and is usually ignored. It arises from the fact that, even in our best schools, boys are never taught the alphabet, and have no knowledge of the true meaning of many of the symbols which it contains. All that they really understand 16 PROCEEDINGS OF THE BRITISH ACADEMY are such consonantal symbols as b, d, and the like, the values of which hardly ever vary. But of the history of the vowels they usually know nothing. For this there is one easy remedy, yet it usually meets with a most stubborn and unintelligent resistance. The remedy is simply this — to teach boys to pronounce Latin properly. As all our symbols are of Latin origin, the first lesson, upon which all other lessons in phonetics are founded, is to learn how Latin was pronounced in classical times. Any one who has once learnt the true sounds of the Latin symbols, a task which may be achieved with sufficient accuracy, not perhaps in a few hours, but at any rate in a few days, by the merest beginner who can read at all, has the true clue to the history of nearly every language in Europe. Yet such is the apathy, such the disregard of this simple yet essential lesson, that, so far from encouraging it, most teachers do their worst in the endeavour to suppress it, and to inculcate the manifest falsehood that the Romans in the time of Augustus employed the pronuncia- tion of the modern Londoner. It has often struck me as a sad satire upon our pretended 'scholarship,' that a student who has ac- quired such a mastery over the technical structure of Latin verse that he is able to compose lines which he believes to be hardly in- ferior to those of Virgil or Ovid, is at the same time often so wholly ignorant of the Latin alphabet that he could not possibly read his own lines in a way that a Roinan poet would have recognized as being intelligible. It is astonishing that our teachers should recoil from so simple a task as that of teaching the alphabet, as if it were an endeavour wholly beyond their strength and ability ! How they can endure to confess such a helpless incompetence is quite unintel- ligible to me; at any rate at the present date, when the old sounds are so well understood. Surely they will not always desire to be behind the age, to refuse to move with the times, and to let all other nations surpass them. This is the conclusion of the whole matter. The writings of Dr. Ellis and Dr. Sweet have given the clue to the meanings of our spellings. We may deduce from them the general rule, that a large number of our words are so spelt^ as to show how they were formerly pronounced, which at once explains why they are no longer phoneti- cally exact. Spelling reform is no new thing, but was frequently considered and adopted in past times ; yet it has always been some- what behind the age, and is now no longer a living principle. This is a "heavy loss from a practical point of view, and has produced much false etymology. A scientific spelling reform, which would adopt the Italian sounds of the vowels as a chief guide, is a thing much desired by all who have studied the subject, and affords the THE PROBLEM OF SPELLING REFORM 17 only adequate solution. Meanwhile, even a partial- refbrimwould prove a great benefit; and the nature of iSuch a partial retJorm can be easily gathered from a, perusal of Dr. Sweet's lists, as drawn up in 1881. There was at that date a slight hope ofi success, but whatever chance it had was promptly howled' down by writers for the daily and weekly press, who did not understand so much as the simple rudiments of the problem, and i had Neither sympathy with nor mercy upon any one who was capable of instructing them. The same unsuspected ignorance prevails still, and is aS^ noisy and contemptuous as ever. The only~ hope obviously hes in the possi- bility that a suspicion of their own ignorance may some day dawn upon writers who, after all, are uniaware of their prejudice and are only unintentionally dishonest. > They cah only • seei the matter in its true light when the study of 'phonetics has advanced much further than at present. The reform 1;hatwould best tend to this desirable result would be the teaching, in our great schools, of a correct Latin pronunciation. This would at ortce draw attention to the fact that our symbols have a history and a meaning ; and the old silly habit of appealing to the e^e as the best judge of ioimds would cease to command any serious .attention or to be ilnfiiirlyi exalted. Before we can hope to amend the written forms weimust, in fact,! learn to understand them ; and the beginning bf such understanding depends upon the sense of the L^tin symbols. It is the relilctancei to learn this that reaUy stops the way. ■ But, at last, there is actually a chance. The Ufnversity of Cam'' bridge many years ago requested- the assistance! of Oxford in prb- posing a scheme for the right pronuricialtion'- of' Latin,' .but was iln- successful. Quite recently i only in' the present year, its applica- tion was renewed, and this time with success. Our two great uni- versities are now prepared to endorse, and in fact have already en- dorsed, a simple and intelligible scheme which, even if it be not per- fectly accurate in minute points, is perfectly intelligible and sufficiently instructive. " If this refonn'iS accepted by the 'schools, there is indeed great hope. The consequences may easily be far- reaching beyond all that might be exp^ed. For the right pronun- ciation of Latin throws a full andckar light not only— as even the ignorant admit — ^upon the sounds and spellings of all the Romance languages, but upon the sounds •and spellings of air periods of every language that has adopted the Roman alphabet, including, indeed, .nearly all the leading languages in Europe," and especially such as concern us most. Viz. Anglo-Saxon, Anglo-French, and the Old Norse. Hitherto, a reformed English spelling ha^ihad merely 18 PROCEEDINGS OF THE BRITISH ACADEMY to be hinted at, in order to be at once howled down and scoffed at ; but a new race of schoolboys would soon arise' who would be ashamed to exhibit such a lamentable mode of argument as that which is solely fourided upon prejudice and derision. It seems to me to be a safe prophecy that, if ever Spelling Reform in Eng- land should come to be an accomplished fact, the beginning of the movement will reaUy date from the present year, 1906, when Ox- ford and Cambridge have agreed, for the first time in recent history, upon a reformed pronunciation of Latin which they are prepared to recommend for the use of every school in the country. If, however, it should come to pass that a real Spelling Reform should previously be effected in America, it may quite possibly be a gain to us ; because the history of thie language is there more gen- erally known. I lately met with the President of an American university, who said to me (I have no doubt with perfect truth) : — 'In our universities English takes the first place.' This is one of those facts of which the ordinary Englishman is entirely ignorant ; indeed, it is almost impossible for him to imagine how such a state of things can be possible. I recommend the contemplation of this astounding fact to your serious consideration. lit 'Mi m 'M R(l^?>'— -.^ I