[Journal of the Simplified Spelling Society, 1987/2 pp17-25 later designated J5.]
[See Journal and Newsletter articles, Pamflet 15 and Cut Spelling by Chris Upward.]
Back to Part 1.

Cut Speling - a Linguistic Universl? Part 2

Christopher Upward.


3.1 Selection of evidnce
3.1.1 Sorces. Th main contrast of English spelings we present here is between presnt-day t.o. and those of th 16-17th centuris (brodly speaking, Shakespearean), befor printrs had acheved a consensus on standrd spelings. But to extend th perspective we also refer to erlir and later periods: to th Midl English of th late 14th century (brodly speaking, Chaucerian), befor printing had been developd in Europ (1453), as wel as briefly to anglo-saxn (c.500-1100 AD) and to th erly 18th century. Three kinds of sorce hav been used. Th first is th Bible, since it is taken to be th standrd english text par exellence in those past centuris, with th most widespred curency and gretst influence, and since it afords th most direct comparisn between th orthografy of difrnt eras (in each case th same section of the Sermn on th Mount has been chosen - Matthew vi, 25-34). Th secnd sorce is litrry, with text-sampls from Chaucer, Shakespeare and Swift. And th third sorce is th word-lists givn in histris of th languaj. A seprat section deals with difrnces between th oldr, presnt forms of british english and th mor advanced websterian forms adoptd in America.

3.1.2 Validity of th evidnce.
In examning th orthograficl evidnce of past ajes befor t.o. had becom standrdized, we hav to be careful not to regard individul texts or words as ofring incontrovertbl proof that can justify absolute, clear-cut conclusions. As D G Scragg [15] demonstrates, th variety of speling (arising from difrnt dialects, difrnt scriptorial traditions and difrnt persnl practices) found particulrly befor th aje of printing makes it imposbl to claim that th sampls givn here ar necesrly typicl of th period in wich they wer put to paper. At best we can hope they ar brodly indicativ of jenrl tendncis, and that global calculations of length (if not individual spelings) do hav som validity. Similrly with th 16-17th century texts: since they do not represent a standrd, we canot make precise comparisns with t.o., but if as a hole they ar al longr, that fact is of som significnce. Wen Scragg (p.71) givs th forms pity, pyty, pitie, pytie, pittie, pyttye al as being current among elizabethan printrs, it is clearly not entirely acurat to say "th elizabethans used a longr speling for pity than t.o. dos"; but we can make jenrl observations such as that most elizabethan spelings of pity wer longr. In al th calculations th letr <y> has been taken as eqivalent of <i>, altho both typograficly and in handriting it is a les economicl symbl.

3.1.3 Presentation of material.
Th texts ar chosen to ilustrate th efect of erlir spelings on continuus ritn english, tho they hav th disadvantaj of limitd and repetitiv vocablry. They hav been reproduced here using as far as posbl th typograficl devices such as th letr yogh and th long <∫> for <s> wich wer found in th copy from wich th text was taken, i.e. if th copy was modrn and did not use th old devices, they hav not been used here. Th word lists on th othr hand provide a wider selection of vocablry, tho many of th words wil hav been less comn.

Th transcriptions sho word-length as folos:
1. bold typ indicates mor letrs than t.o.
2. italics indicate fewr letrs than t.o.
3. undrlined words or letrs indicate a CS form.

Words of th same length as t.o. (tho th speling may difr), or wich do not ocur in t.o., ar printd normly.

3.2 Midl english.
Wyclifs Bible translation (dated 1389) [16]
Th letr yogh <ʒ> may usualy be red as a precursr of <y>, as in ye, you, but in riʒt it forshados <gh>.

25. Therfore Y say to ʒou, that ʒe ben nat besie to ʒoure lijf, what ʒe shulen ete; othir to ʒoure body, with what ʒe shuln be clothid. Wher ʒoure lijf is nat more than mete, and the body more than clothe?

26. Beholde ʒe the fleeʒing foulis of the eir, for thei sowen nat, ne repyn, neither gadren in to bernys; and ʒoure fadir of heuen fedith hem. Wher ʒe ben nat more worthi than thei?

27. Sothely who of ou thenkinge may putte to his stature oo cubite?

28. And of clothing what ben ʒe besye? Beholde ʒe the lilies of the feelde, how thei wexen. Thei trauellen nat, nether spynnen;

29. Trewly I say to ʒou, for whi neither Salamon in al his glorie was keuerid as oon of these.

30. For ʒif God clothith thus the heye of the feeld, that to day is, and to morwe is sente in to the fourneyse, how moche more ʒou of litil feith?

31. Therfore nyl ʒe be bisie, sayinge, What shulen we ete? or, What shulen we drynke? or, With what thing shulen we be keuered?

32. Forsothe heithen men sechen alle these thingis; trewly ʒoure fadir wote that ʒe han need to alle these thingis.

33. Therfore seke ʒee first the kyngdam of God and his riʒtwisnesse, and alle these thingis shulen be cast to ʒou.

34. Therfore nyle ʒe be besie in to the morwe, for the morew day shal be besie to it self; sothely, it sufficith to the day his malice.

Ignoring th problms of comparing this East Midlands orthografy with t.o. wich result especialy from fonolojicl and morfolojicl chanjes, we note 31 cases of aditionl final <e> (beholde), 13 cases of aditionl final <n> or <en> (sowen), 4 cases of aditionl medial vowls (thingis), 3 cases of consonnt dubling (putte), and a few othr extra letrs. On th othr hand som forms ar shortr than in modrn english, especialy wen a t.o. digraf is representd by a singe letr, as with yogh for <gh>, <d> for <th> (fadir), som shortr vowls (as in fedeth, heuen), and singl consonnts insted of dubl (litil, morew). Four forms in fact anticipate CS: al, shal, ther, wher. A few forms wud also look wel in a foneticized modrn english orthografy: thei, feith, worthi. Overal, compared with th t.o. forms, about three times as many words ar longr as ar shortr.

As a litrry sampl of Midl English, we here giv th opening to the Prolog to Chaucer's Canterbury Tales: [l7]


Whan that Aprill with his shoures soote
The droghte of March hath perced to the roote,
And bathed every veyne in swich licour
Of which vertu engendred is the flour;
When Zephirus eek with his sweete breeth
Inspired hath in every holt and heeth
The tendre croppes, and the yonge sonne
Hath in the Ram his halve cours yronne,
And smale foweles maken melodye,
That slepen al the nyght with open ye
(so priketh hem nature in hir corages);
Thanne longen folk to goon on pilgrimages,
And palmeres for to seken straunge strondes,
To ferne halwes, kowthe in sondry londes;
And specially from every shires ende
Of Engelond to Caunterbury they wende,
The hooly blisful martir for to seke,
That hem hath holpen when that they were seeke.

These 18 lines of Londn orthografy merely serv as an ilustration, and th figrs givn belo for relativ word-length covr th first 100 lines. Decisions had to be taken on sevrl points of uncertnty. For exampl, th verb-inflection <-(e)th>, as in hath, bygynneth, was countd as a longr speling of th presnt inflection <-s>, rathr than as a form with no t.o. eqivalent, since insofar as hath (wich ocurs very freqently) uses 4 letrs to spel 3 fonemes, the 3-letr modrn form is grafemicly mor economicl. But by th same lojic, we then hav to say that th forms hem, hir ar mor economicl than ther modrn eqivalents them, their. With gramaticl forms like foughten, wich ar both fonemicly and orthograficly longr than th modrn english fought, shortning has ocurd in both respects. Chaucer's languaj being that of pre-gret-vowl-shift english, we also find spelings like lyf, wys that reflect th pronunciation befor th presnt long vowl was introduced; in one sense, th t.o. forms life, wise represent a lengthning since Chaucer's day, but in anothr sense, th modrn speling merely reflects th modern vowl-length, and to talk of th modrn speling being less economicl is perhaps not meaningfl; nevrthles Chaucer's forms hav been countd as shortr, just as Caunterbury has been countd as longr, and just as we noted th pre-vowl-shift forms of Midl Hy Jermn as shortr than ther modrn eqivalents. Th results wer as folos: with th title, th 100 lines contain 755 words, of wich 545 wer eithr th same length as ther t.o. eqivalent, or els had no direct modrn desendnt; 133 wer longr than th modrn eqivalent, and 77 shortr. Of those that wer shortr, 26 wer precisely CS forms (notably, repeatd ocurences of al, wel, ful, ther), wile 16 othr words containd CS featurs (notably simplifyd dubld consonnts in maner, riden, litel, acordaunt).

3.3 16th & 17th century english.
Tyndale's Bible Translation(1526)
We here giv th same verses from St Matthew as in th Wyclif translation. A direct comparisn between them is dificlt because Wyclif translated secnd-hand from latin, wile Tyndale had access to th orijnl greek.

25. Therefore I saye vnto you, be not carefull for youre lyfe what ye shall eate, or what ye shall dryncke; nor yet for youre boddy, what rayment ye shall weare. Ys not the lyfe more worth then meate, and the boddy more off value then rayment?

26. Beholde the foules of the aier, for they sowe not, neder reepe, nor yet cary into the barnes; and yett youre hevenly father fedeth them. Are ye not better then they?

27. Whiche off you though he toke taught therefore coulde put one cubit vnto his stature?

28. And why care ye then for rayment? Beholde the lyles off the felde, howe thy growe. They labour not nether spynn;

29. And yet for all that I saie vnto you, that even Solomon in all his royalte was nott arayed lyke vnto one of these.

30. Wherfore yf God so clothe the grasse, which ys to daye in the felde, and to morowe shalbe cast into the fournace, shall he not moche more do the same vnto you, o ye off lytle fayth?

31. Therfore take no thought, saynge, What shall we eate? or, What shall we dryncke? or, Wherewith shall we be clothed?

32. Aftre all these thynges seke the gentyls; for youre hevenly father knoweth that ye have neade off all these thynges.

33. But rather seke ye fyrst the kyngdom off heven and the rightewesnes ther of, and all these thynges shalbe ministred vnto you.

34. Care not therfore for the daye foloynge, for the daye foloynge shall care ffor yt sylfe; eche dayes trouble ys sufficient for the saine silfe day.

In this version we find 57 words longr than in t.o., and 22 shortr (not counting ye), of wich 14 hav at least som featur of CS. Inconsistncy is particulrly markd.

The Bishops Bible (1588)
The Gospel by ∫aint Matthæwe

25. Therefore I ∫ay vnto you, Be not careful for your lyfe, what ye ∫hal eate or drinke, nor yet for your body what ye ∫hal put on: Is not the life more woorth then meate; and the body then rayment?

26. Beholde the fowles of the ayre: for they ∫owe not, neyther doo they reape, nor carrie into barnes, yet your heauenly father feedeth them. Are ye not muche better then they?

27. Which of you by taking of careful thought can adde one cubite vnto his ∫tature?

28. And why care ye for rayment? Learne of the Lylies of the fielde, how they grow; they weery not (them∫elues) with labour: neyther (doo they) ∫pinne.

29. And yet I ∫ay vnto you that euen Solomon in al his royaltie, was not arayed lyke one of the∫e.

30. Wherefore, if GOD ∫o clothe the gra∫∫e of the fielde, whiche though it ∫tande to day, is tomorowe ca∫te into the Ouen: ∫hal he not muche more (doo) the ∫ame for you, (O) ye of litle fayth?

31. Therefore take no thought, ∫ayeing, what ∫hal we eate? or, what ∫hal we drynke? or, wherwith ∫hal we be clothed?

32. (For after al the∫e thynges doo the Gentiles ∫eeke:) for your heauenlye father knoweth that ye haue neede of al the∫e thynges.

33. But ∫eeke ye fyr∫te the kyngdome of God, and his righteou∫ne∫∫e, and al the∫e thynges ∫halbe added vnto you.

34. Care not then for the morowe: for the morowe ∫hal care for it ∫elfe: Sufficient vnto the day, is the euyl therof.

Th speling here is much closer than Tyndale's to t.o.: altho ther ar actuly 37 final <e>s not presnt in t.o. (Tyndale had 35), ther is only one dubld consonnt not in t.o. (spinne), wher Tyndale had 15; like Tyndale, the Bishops Bible has 14 CS forms, of wich 8 ar shal. One word has been relengthnd from Tyndale: heven has becom heauen). A new peculiarity is th speling doo. Overall, 46 words ar longr than in t.o., and 15 shortr.

The Authrized Version (1611).
This version moves yet nearr to t.o., with only 19 final <e>s not found in t.o., and exept for th compound shalbe <l> is usuly dubld as in t.o. (and aditionly in wherewithall, lillies, evill). Th only othr CS forms remaining ar therfore, arayed. Th Authrized Version dos howevr tend to lengthn we, ye to wee, yee in a manr not found in erlir versions, and th freqency of yee for ye and shall for shal means that insofar as the Bishops Bible and the Authrized Version hav th same wording, th latr is 5 letrs longr. For th first time we meet th t.o. form little. Of th 251 words of this extract from th Authrized Version only 4 hav fewr letrs than t.o., wile 31 words hav mor (these totals include repeatd words).

We here giv th speling of th First Folio edition (1623) of th To be or not to be soliloqy in Hamlet.

To be, or not to be, that is the Que∫tion:
Whether 'tis Nobler in the minde to ∫uffer
The Slings and Arrowes of outragious Fortune,
Or to take Armes again∫t a Sea of troubles,
And by oppo∫ing end them: to dye, to ∫leepe
No more; and by a ∫leepe, to ∫ay we end
The Heart-ake, and the thou∫and naturall ∫hockes
That Fle∫h is heyre too? 'Tis a con∫ummation
Deuoutly to be wi∫h'd. To dye to ∫leepe,
To ∫leepe, perchance to Dreame; I, there's the rub,
For in that ∫leepe of death, what dreames may come,
When we haue ∫huffel'd off this mortall coile,
Must giue vs paw∫e. There's the re∫pect
That makes Calamity of ∫o long life:
For who would beare the Whips and Scornes of time,
The Oppre∫∫ors wrong, the poore mans Contumely,
The pangs of di∫priz'd Loue, the Lawes delay,
The in∫olence of Office, and the Spurnes
That patient merit of the vnworthy takes,
When he him∫elfe might his Quietus make
With a bare Bodkin? Who would these Fardles beare
To grunt and ∫weat vnder a weary life,
But that the dread of ∫omething after death,
The vndi∫couered Countrey, from who∫e Borne
No Traueller returnes, Puzels the will,
And makes vs rather beare tho∫e illes we haue,
Then flye to others that we know not of.
Thus Con∫cience does make Cowards of vs all,
And thus the Natiue hew of Re∫olution
Is ∫icklied o're, with the pale ca∫t of Thought,
And enterprizes of great pith and moment,
With this regard their Currants turn away,
And loo∫e the name of Action. Soft you now,
The faire Ophelia? Nimph, in thy Orizons
Be all my ∫innes remembred.

Of th 275 words in this text, 32 ar longr than in t.o., sinnes being 2 letrs longr; 12 words ar difrntly spelt but of th same length (disregarding th <u, v> altrnations); and only 3, ake, puzels, I (and perhaps remembred) ar shortr, tho th absnce of apostrofes to indicate posession (the Lawes delay) shortns 3 mor words. (Apostrofes ar howevr used for omission: there's). No CS forms occur, unless we take th secnd <r> in remembred as sylabografic.

3.4 Erly 18th century.
Jonathan Swift A Modest Proposal (1729).
Altho John Wilkins' 1668 reform proposal (Scragg p. 101) includes such obvius CS forms as erth, giv, bred, gost, ded, shal, its radicl incorporation of diacritics and greek letrs ruled it out as a practicl proposition. By th late 17th century a consensus among printrs had developd wich amountd to a larjly standrd orthografy and was substantialy wat we no as t.o., tho certn featurs had yet to be establishd, th most obvius (if superficial) deviations from t.o. being th survival of th long <∫> and th much mor librl use of capitl letrs. An analysis of Jonathan Swift's pamflet, A Modest Proposal, givs a fair impression of othr spelings that difrd from th futur t.o. Th 10 pajes contain 25 words with longr spelings than t.o., 30 words with non-t.o. forms of th same length, 4 words hose form is shortr than t.o., plus anothr 3 or 4 doutful cases (e.g. missing apostrofes). Of th 25 longr words, most hav extra consonnt-letrs: 9 end in <-ick> wher t.o. has <-ic>: publick (7), prolifick, catholick; 6 hav a hyfn or space wher t.o. dos not: common-wealth, play-house, over-run, our selves, my self, land-lords; and 3 hav dubld <l> wher t.o. has it singl: melancholly, tollerably, skillful. Th only longr vowl-spelling is cloathing. Th forms having th same length as t.o., on th othr hand, mostly deviate in vowl-speling, with sevrl patrns aparent: <y> for <i> in 9 words: likelyhood, dye, lyable (2), boyled, dyet, supplyed, crucifyed, tythes; <e> for <i> or vice versa in th unstressd initial sylabl of 7 words: encrease (3), imploy (tho employ ocurs mor ofn), degressed, incumbrance, intailing; <c> for <s> in 5 words: expence (4), parcimony; and <-ance> for <-ence> in consistance, subsistance. Th shortr words ar stroling, dropt, fricasie, inclemences. Clearly this text, wich was publishd a century aftr Shakespeare's first folio, has progressd a long way towards t.o. (only som 60 words over 10 pajes hav non-t.o. forms, wheras in th To be or not to be speech nearly one word in 5 had yet to acheve th t.o. form); but a major continuing source of deviation is th failur to cut redundnt letrs that t.o. later removed.

3.5 Discussion of th texts.
Th abov texts demonstrate abov al inconsistncy of speling, both regarding choice of letrs, and in th numbr used. Not econmy, but with th advent of printing typograficl reqiremnts ofn determind th numbr of letrs, th compositr using mor or fewr to complete a line or to fit words into his justifyd text (see Scragg Ch. 5). Th extra letrs wer predominntly <e> and dubld consonnts, wich Scragg analyses as folos: "writers used final <e> in a quite haphazard way; in printed books of the sixteenth century <e> was added to almost every word which would otherwise end in a single consonant, though the fact that it was then apparently felt necessary to indicate a short stem vowel by doubling the consonant (e.g. bedde, cumme, fludde) shows that writers already felt that final <e> otherwise indicated a long stem vowel." That som elizabethans wer aware of this superfluity of letrs, and indeed of th CS principle as such, is shown by Claude Desainliens remark [18] in 1576 concerning "the great strife betwene them that woulde have our tongue written after the auncient orthographie, and those that do take away many letters as superfluous in writing". Howevr randm th variations our sampl of texts contains, th avraj printd word had mor letrs than its t.o. form. But was this tru of handriting too? Scragg qotes (p.68-69) from a letr by Qeen Elizabeth I (1586), wich tho inconsistnt and speling 19 out of 116 words with an <e> not found in t.o., nevrthles shortns as many words as it lengthns (13) and uses sevrl ful CS forms: wer, wel, dout, and sevrl partial ones: counselar, shuld, folowe. Likewise Pitman's list [19] of 34 variants on ther surname used by membrs of Shakespeare's famly contains 24 shortr forms (Shaxper is th shortst), 8 forms of equal length, but only 2 longr ones. This sujests th intriging posbility that privat persns used mor economicl spelings than did printrs with ther typograficl constraints - a hypothesis wich a wider survey of elizabethan manuscripts cud esily verify or refute. If it proved to be th case, it wud imply that printing tecniqes wer to blame for many of th cumbrsm and iregulr forms of t.o., and that riters left to ther own natrl inclinations wud long ago hav taken th Cut Speling road.

3.6 Shortning of individual words.
We alredy noted (§2.5) that som jermn spelings became shortr as a result of syncope. Th same fenomenn ocurd in english, with th spoken form of a word losing sylabls or fonemes, and th ritn form contracting to mach; a strikingly aposit exampl is copse (it derives from french couper 'to cut') hose longr form coppice stil exists. Th Oxford Dictionary of Place-Names [20] ofrs a masivly comprehensiv colection of exampls, mostly going bak to anglo-saxn times; th first 20 names it lists (excluding celtic Abergavenny) reqire 156 letrs today, wheras ther erly forms reqired 200, th two most extreme cases being Abberton (Worcs), ritn Eadbrihtincgtun in 972, and Abson (Glos.), ritn Abbedeston in 1167. Similr streamlining is obtaind wen CS is aplyd to som modrn place names, such as Lestr, Birmingm, Londn. Heller [21] givs som striking exampls of th same process afecting ordnry words: bryd-ealu>bridal, clerec>clerk, conestable>constable, enaleofon>eleven, furhlang>furlong, God be with you>goodbye, gose-somer>gossamer, hengestmann>henchmen, Magdelin>maudlin. Strang [22] givs a mor systmatic acount of th way many anglo-saxn forms hav shortnd thru th centuris; her exampls include: cylen>kiln, ealle>all, eornostlice>earnestly, godspel>gospel, heafod>head, heall>hall, hlaf>loaf, hlaford>lord, makede>made, munuc>monk, mynet>mint, saternesdæg>Saturday, sweostor>sister. Mor modrn exampls (som of them exemplifying aphesis rathr than syncope) ar: escheat>cheat, keverchef>kerchief, moneth>month, napron>apron, perambulator>pram, omnibus>bus, caravan>van (and soon telephone>fone?). Such shortnings ar good modls for CS forms like busness, cupbrd, hansm, histry, poisnus, rasbery, sovrn, wensday. But ther hav been othr patrns of shortning wich do not lend themselvs so obviusly for CS: th loss of th anglo-saxn inflections has ment that for instnce verbs of anglo-saxn orijn hav lost endings, as singan>sing (tho som redundnt final <e>s in t.o., as in have, love etc, can be similrly shortnd in CS); for purely practicl, semantic reasns inflammable has recently reduced to flammable; and english is wel-nown for verbl contractions like they'll, we've, wouldn't, won't in wich th exeedingly comn spoken form has its acceptd (if somwat informl) ritn eqivalent.

3.7 Lengthning of individul words.
Alongside al these cases of fonolojicl and conseqent orthografic shortning, english has also seen varius patrns of orthografic lengthning. One categry involvs th replacemnt of a singl letr by a digraf, as wen with th advent of printing thorn and eth finaly gave way to <th> and pre-consonantl yogh gave way to <gh>; between <th> and <d> we find both lengthning and shortning: modor>mother, fadr>father, but murther>murder, burthen>burden; th use of th 'french' digraf <ch> aftr th normn invasion gave rise to such lengthnings as cest>chest; vowl-digrafs wer introduced to reflect vowl-lengthning, as boc>book, dohtor>daughter, fyr>fire, mus>twuse, swin>swine. (Similr patrns of lengthning ar proposed by such speling-reform scemes as New Spelling [23], wich wud substitute <ks> for <x>, and rite tuberculosis as tuebercueloesis.) An intresting case is anglo-saxn æpl wich lengthnd to t.o. apple, but wich CS wud cut bak to apl.

A completely difrnt categry consists of words hose speling was delibratly lengthnd for suposedly etymolojicl reasns, and hose pronunciation was then somtimes adaptd to mach (speling-pronunciation). Scragg (p.57-59) lists th foloing: perfit>perfect, assoil>absolve, amonest>admonish, caitif> captive, cors>corpse, langage>language, trone>throne, cedule>schedule, samon>salmon, ceptre>sceptre, teatre> theatre. Ocasionly a word was shortnd despite th adition of a letr: dette>debt, wile doute>doubt kept th same length. One word was lengthnd, only to be shortnd again centuris later: fantasy>phantasy>fantasy. A good few aditionl letrs wer based on mistaken etymology: sithe>scythe, yland>island, avauncen>advance, avauntage>advantage, avice>advice, amiral>admiral, ancre>anchor; one of these was shortnd wile being lengthnd: emeraude>emerald, wile auctor kept its length as author. Could aqired its <l> by false analojy with would, and whole, whore, whoop aqired <w> to distinguish them from homofones.

3.8 Lengthning v. shortning: discussion.
Wile we can describe th shortning of spelings to mach ther reduced fonolojicl length as a natrl and lojicl process, many of th abov cases of lengthning ar far from natrl or lojicl, being rathr an eruption of linguistic erudition wich, even wen not based on totaly false premises, has litl to do with facilitating comunication and litracy. Shakespeare satirized it in th caractr of Holofernes (suposedly based on th elizabethan orthografr Mulcaster) in Love's Labour's Lost (V, 1), ho is made to say (here in t.o.) : "I abhor... such rackers of orthography, as to speak dout, when he should say doubt; det, when he should pronounce debt, - d,e,b,t, not d,e,t... This is abhominable, which he would call abominable." We hav alredy qoted Desainliens, and we may here repeat Scragg's qotation (p.98) from Coote's 1596 statemnt of th CS principl: "I haue differed in writing many syllables from the vsual manner, as templ without (e), tun with one (n) and plums not plummes etc; my reason is, I haue put no moe letters than are of absolute necessities.." Thus th elizabethans themselvs wer clearly aware of th perverseness of many lengthnings, and as both Scragg and our analysis of th Swift text (§3.4) sho, aftr th elizabethan aje most speling chanjes involvd cuting: th silent final <e> larjly disapeard unless it servd to indicate a preceding long vowl, th <ie> ending as in academie became <y>, th <-ick> ending as in musick lost its <k>, controul became control. But altho th silent final <e> in words like are, were, there, done, gone was ofn dropd bak in th 16th century, in those comn words, and in othrs like give, live, love, t.o. has retaind it. Similrly th comn simplification of <l> in al, ful, wel etc. 300 years and mor ago has not been adoptd in t.o. (exept in compounds like also, fulfil, welcome). In such cases CS is proposing fidelity to simplr, oldr forms that t.o. has thotlessly inflated.

3.9 Americn shortr than british. [24] [25]
Those americn spelings that difr from those custmry in Britn today wer mostly fixd at a later date than we hav so far discussd, being du to th americn lexicografr Noah Webster, ho put forwrd qite radicl reform proposals in 1789, and then progressivly watrd them down in successiv editions of his dictionry in 1806, 1828, 1838. If americn english is regardd as less conservativ than british english, that difrnce is widely reflectd in th mor economicl speling, particulrly in racy forms such as tonite, tho, thru wich hav not howevr yet acheved oficial status. Among th fuly acceptd difrnces, we find that british <ou> in mould, moult and in th <-our> ending becoms <o> (mold, molt, labor); british <ae, oe> ar ofn plain <e> (esthetic, maneuver); verbs ending in an unstresd sylabl with a final consonnt ofn don't dubl th consonnt befor inflections (worshipd rather than worshipped - compare also woolen, wagon for woollen, waggon); and final <-e> is mor ofn dropd than in british english wen a foloing sufix begins with a vowl (milage, salable as against mileage, saleable); silent <ue> endings disapear as in catalog, and a programme becoms a program; some tools hav shortnd ther handls (ax, adz); and som <gh, ph> digrafs hav been rationlized, as in boro, plow, draft, sulfur. An isolated case is US check for british cheque. Th shortr arnericn forms ar not always just a matr of speling, but ocasionly reflect th 'shortr' americn pronunciation (mustache, aluminum as against moustache, aluminium). Som americn forms hav taken th prefrnce for <i> over <y> or <e> furthr than british (compare Swift) as in inclose, tire, stories for enclose, tyre, storeys. Americn may also economize by using fewr hyfns than british, as cooperate rathr than co-operate. Scragg notes (p.85) that som shortr arnericn forms ar incresingly used in th british press. Ther are howevr also a few americn spelings wich ar no shortr than th british form (center, defense, gray), and a few british forms ar shortr than th americn, particulrly wher a dubld consonnt is involvd: US biassed, skillful, fulfill, benefitted as against british biased, skilful, fulfil, benefited.


4.1 Linguistic, sycolojicl, economic.
To describe CS as "biolojicly necesry" is clearly to overstate th case, but in this articl we hav asembld evidnce that th omission of superfluus letrs, caractrs, symbls, marks, diacritics or grafemes has in th past and across many languajs been a prime tecniqe of rationl reform. Usuly th omissions relate to th internl reqiremnts of th riting systm, as wen they reflect fonolojicl simplification, or th fact that morfolojicl distinctions made in erlir forms of th ritn languaj wer later no longr felt to be necesry or helpful. But it is not unreasnbl to sujest that in adition to such strictly linguistic motivations, powrful lojicl, sycolojicl and economic motivations ar also at work. Cal it laziness, or cal it eficiency, human beings wil natrly wish to carry out ther apointd tasks with th minimm of efrt (th mor economicl elizabethan handriting, and th freqency with wich speling mistakes shortn words, ar furthr indications of such a tendncy wich need exploring). Th basic tasks imposed by a riting systm ar firstly lerning and secndly using, and compactness clearly represents an importnt component in a systm desynd for th eficient performnce of both tasks, especialy wen compactness reflects systemic simplicity as wel. But th sycolojicl atraction of simplicity is reinforced by th potential economic and financial benefits. Wethr or not a global, carefuly pland speling reform is introduced for english, economic pressurs wil al be operating in th direction of CS. As Scragg says (p.85-86): "Printers in particular are attracted by the appreciable savings in production costs to be made by the use of spellings which economise in space... it is likely that publishers will be unable ultimately to resist the saving in paper, ink, and type-setting labour involved in the shortened forms... Brevity is the keynote of present developments in spelling. Fully established in the style-sheets now are loth against loath, curtsy against curtsey, hiccup against hiccough ... against th Daily Express dulness and fulness, The Guardian has dullness and fullness."

4.2 CS as a natrl reform stratejy.
CS also has to be considrd in terms of practicality of implementation. Major upheavls in riting systms ar rare, and wen they hav taken place, we note th combination of a larjly ilitrat society and an authoritarian rejime, as wen Turky replaced arabic script with th roman alfabet in 1928, or wen th Soviet Union convertd azerbaijani from arabic script to roman in 1924, and then to cyrilic in 1940. [26] By contrast, in th many societis of th english-speaking world wich ar culturaly sofisticated and mor or les fuly litrat, with hyly developd publishing industris and educationl systms, not to mention public opinion as a crucial influence on decision-makers, a radicl upheavl cud not be countnnced. Th only feasbl reform wud be one that capitlized on th natrl processs and tendncis we hav observd operating in othr languajs and at erlir periods of english. These processs and tendncis ar caractrized by gradualness, wich ensures th mutual intelijbility of old and new systms, so that wile childrn and advanced publishrs use th new spelings, th oldr jeneration and mor conservativly mindd peple can stil use th old. English reqires new spelings wich readrs and riters wil accept as redily as today we accept such radicly difrnt forms as gaol, jail. Bernard Shaw, Sir James Pitman and John Downing hav al drawn a paralel with th coexistnce of roman and arabic numerals: each has a role, but just as arabic numerals ar by far th mor eficient for al practicl purposes of calculation, so if CS and t.o. existd side by side for a transitionl period, ther can be litl dout wich wud soon be preferd for most practicl purposes of eficient comunication. As part of our argumnt in favor of speling reform, we must be able not just to present an orthografy that we can demonstrate is systemicly superir to t.o.; we also hav to sho that it fits smoothly into norml historicl and social processs. Th evidnce presentd in this articl surely sujests that by its very natur CS fits th bil.


[15] D G Scragg A history of English spelling, Manchester University Press, 1974.

[16] Joseph Bosworth The Gothic and Anglo-Saxon Gospels in Parallel Columns with the Versions of Wycliffe and Tyndale, London: John Russell Smith, 1865, pp.27, 29.

[17] Geoffrey Chaucer Canterbury Tales, ed. A C Cawley, London: J M Dent & Sons, 1958, reprinted 1981.

[18] quoted in Bror Danielsson John Hart's Works on English Orthography and Pronunciation p.65.

[19] Sir James Pitman & John St John Alphabets and Reading, London: Sir Isaac Pitman & Sons Ltd., 1969, p.65.

[20] Eilert Ekwall The Concise Oxford Dictionary of English Place-Names, Oxford University Press, 4th edition 1960

[21] Louis Heller, Alexander Humez, Malcah Dror The Private Lives of English Words, London: Routledge & Kegan Paul, 1984.

[22] Barbara M H Strang A History of English, London: Methuen, 1970.

[23] Walter Ripman & William Archer New Spelling, London: Sir Isaac Pitman & Sons, 1948, 6th edition revised by Daniel Jones & Harold Orton.

[24] The Oxford Dictionary for Writers and Editors, Oxford: Clarendon Press, 1981, reprinted 1984, p.424.

[25] Robert Ilson 'Diversity in Unity: American and British English' in English Today October 1985, pp.7-11.

[26] Kenneth Katzner The Languages of the World, London: Routledge & Kegan Paul, 1986, pp.156-7, 123.

Back to the top.