Webster as an orthographic conservative?

« previous post | next post »

Matthew Edney, who describes himself as "a British-born academic who now, 27 years after first arrival, is linguistically located somewhere in the general confusion of the mid-Atlantic", sent me an interesting query about the history of English spelling. Since I know almost nothing about this subject, I'm forwarding the question to LL readers, who are likely among them to have the answers, or at least some useful observations.

My query is prompted by an article in today's Guardian: Sarah Churchwell, "English: it's a neologism thang, innit", 5/9/2010.

In the midst of commenting on the inclusion of "thang" etc. in the official Scrabble dictionary, the commentator (an American) makes reference to how Noah Webster nationalized American spelling (-ize rather than -ise, -or rather than -our, etc.). I have seen this opinion in many places, from Bill Bryson's history of the English language to other online journalists.

BUT, my own research suggests a different historical process. In studying the British mapping of India between 1760 and 1840 for my dissertation, I observed a coherent shift in the written language of the officers of the East India Company, in particular from almost universal -or in the 1760s to universal -our by the 1830s. That is, the divergence was not one of US nationalism — Webster only described and codified long-standing linguistic practice in the New England colonies (prescribing them for other parts of the nascent USA). Rather the divergence was on the other side of the Atlantic, as the middle and upper-classes in Britain got all snooty and Frenchified their spelling (which is ironic, given that the era of change was primarily taken up in the long wars against the French).

So, my query is actually twofold:

a) am I off base in my identification of what seems to be an historical myth of the origins of American English as one more attempt to justify the exceptional nature of American culture?

b) if I am correct, what accounted for the changes in British spelling during the period 1780-1830?

[Above is a guest query by Mathew Edney.]

My uninformed reaction is that -our for -or doesn't seem very much like early-19th-century Frenchification, since the cognate words en français at that time ended in -eur.  And as a small empirical contribution, I note that the 1776 edition of Gibbon's Decline and Fall has 20 instances of "splendor" but 14 of "splendour"; 94 instances of "valour" and none of "valor"; 10 instances of "colour" and none of "color"; 40 instances of "vigour" and none of "vigor"; etc.

As for -ize/-ise, the OED agrees that the Frenchification concept applies:

… in modern French the suffix has become -iser, alike in words from Greek, as baptiserévangéliserorganiser, and those formed after them from Latin, as civilisercicatriserhumaniser. Hence, some have used the spelling -ise in English, as in French, for all these words, and some prefer -ise in words formed in French or English from Latin elements, retaining -ize for those formed < Greek elements. But the suffix itself, whatever the element to which it is added, is in its origin the Greek -ιζειν, Latin -izāre; and, as the pronunciation is also with z, there is no reason why in English the special French spelling should be followed, in opposition to that which is at once etymological and phonetic. In this Dictionary the termination is uniformly written -ize.

But as this culturomics graph for British books suggests, the process of Frenchification was apparently not fully established in British practice until the end of the 19th century:


  1. John said,

    May 10, 2011 @ 9:45 am

    Of course, picking the OED as the example is bound to favour (!) the -ize spelling, because that's the Oxford style. British usage generally, though, tends to favour -ise, and the Cambridge Guide to English Usage (which I happen to have to hand) has a more balanced discussion of the -ize/-ise situation, including the problems with appealing to classical roots to justify -ize as a universal English spelling.

    One main argument used to defend -ise spellings in the CGEU is that there are far fewer exceptions if you default to -ise than there are if you default to -ize. But the geographical usage argument tends to make -ize preferable if you're writing for an international audience.

    [(myl) But the fact that -ize is "Oxford style" — and has been for some time — does tend to undermine the story that -ize was a Webster "reform". (Though I know nothing about whether opinion at Oxford on this question had crystallized before Webster or not…)]

  2. Craig Russell said,

    May 10, 2011 @ 9:47 am

    The OED entry on "colo(u)r" identifies the -our spelling with "late Anglo-Norman French" (this also explains the first vowel: colour vs modern French couleur). It says that this (colour) has been "the normal English spelling" since the 14th century, and its examples seem to bear this out (with a few other early spellings: "colur", "colowers", "colors"). "Colour" (based on their examples) is standard long before 1760.

    At least for the word colo(u)r, it seems unlikely that -our is a result of *late* Frenchification, since the ending corresponds to medieval French spelling. It does seem possible that during this period there was a movement to regularize the spelling on analogy with words like colour. OED's citations for splendo(u)r, for example, show far greater variation (splendour, splendeur, splendor) in all periods than "colour".

    I would guess (not being at all an expert) that the period between 1780-1830 was one in which there was already a general movement towards regularization of spelling in many ways, and that if this is when the British -our becomes more prominent, it is only because writers are increasingly attentive to consistency in spelling and felt that ending some Latinate words in -our and some in -or was messy, so they might as well follow the example of (e.g.) colour for everything.

  3. J. W. Brewer said,

    May 10, 2011 @ 10:24 am

    The original 1787 text of the U.S. Constitution has "Labour" (in a now-obsolete clause referring euphemistically to fugitive slaves). But it also has "Offenses" where the Declaration of Independence had "Offences" barely a decade earlier (although the Constitution lacks internal consistency on the -ce/-se issue by using "defence" instead of "defense"). For another example of orthographic change in progress, the 12th Amendment (ratified 1804) already had "choose" instead of the 1787 original's "chuse." If there was regional variation as suggested above (with Webster reflecting specific New England conventions), I don't know what sort of "national" style one should have expected for documents approved by delegates from all up and down the Atlantic seabord.

  4. Dimitri said,

    May 10, 2011 @ 10:30 am

    I wonder why it is that Latin has had such a strong influence on grammatical rules (e.g. no split infinitives) but not on orthographic conventions. Surely well-educated English speakers prior to the 20th century would have known much more Latin than most everyone and could have modeled their spelling after the Latin, at least for obvious words like labor, color, etc.

  5. dw said,

    May 10, 2011 @ 10:32 am

    The Wikipedia article on this subject is quite good.

    The British preference for -ise over -ize is quite recent. Johnson recommended not only -ize in words such as "realize" but also -yze in "analyze" etc. My sense is that it began in the late nineteenth century, when British spelling was becoming Frenchified (see also "programme" versus "program"), and gradually hardened during the twentieth as a reaction to American cultural hegemony.

    The -our/-or question is of a somewhat different nature. Here it probably is fair to credit/blame Webster with the American preference for -or. However, there are many words formerly spelled with -our that are now universally spelled -or, even in Britain. This includes all words where the ending is preceded by , as in erro(u)r, mirro(u)r, etc.

  6. J. W. Brewer said,

    May 10, 2011 @ 10:33 am

    Emily Post's 1922 advice on the wording of wedding invitations says: The invitation to the ceremony should always request “the honour” of your “presence,” and never the “pleasure” of your “company.” (Honour is spelled in the old-fashioned way, with a “u” instead of “honor.”)

    Of course, that she perceived/described the -our variant as "old-fashioned" doesn't make it so; pretentious Anglophilia and pretentious archaism can merge into each other given an unreflective (and often empirically inaccurate) implicit premise that the Mother Country must do it the Old-Fashioned Way.

    It may be significant here that until well into the 20th century most Americans would have been exposed to the King James Version which was generally, if not universally printed in U.S. editions without having its spelling Websterized (although it typically reflects spellings that stabilized in the mid-18th C. rather than those of the 1611 original edition — for reasons I've never understood the only c. 1600 English texts that seem to be consistently published with original spelling are the poems of Spenser). So even if the orthographic choices of the post-1762(?) editions of the KJV did not accurately reflect uniform earlier BrE practice, they were probably taken by Americans to do so to the extent they varied form their own Websterized usage.

  7. Barney said,

    May 10, 2011 @ 11:12 am

    Dimitri: The split infinitive is not not ungrammatical in English, and apparently never has been in the last 700 years. There's a widespread and persistent myth that it is ungrammatical. Split infinitives are impossible in Latin, not just ungrammatical.

  8. Morten Jonsson said,

    May 10, 2011 @ 11:19 am

    The reason Spenser tends not to be modernized is that he spelled that way on purpose–he deliberately used words and spellings that were archaic in the 1580s. Whereas Shakespeare and that crowd spelled the way they did because they didn't know any better. That seems to be the thinking, anyway.

    An interesting case is "glamour." It's not French, of course, but Gaelic, and something like "glammer" would be truer to the derivation. It was Sir Walter Scott who popularized the -our spelling, which probably looked right to him because that was the prevalent spelling by then in words that actually were French. And his spelling has stuck, for the most part, on both sides of the Atlantic, because it would be silly to change it to an -or that never really existed (not that silliness has ever been an impediment to mandating spelling).

  9. E W Gilman said,

    May 10, 2011 @ 11:43 am

    My recollection is that none of the spellings that Webster preferred (-or for -our, -ize for -ise, -er for -re) in his spelling books and dictionaries was an innovation, but rather was based on older forms. He seems basically to have been trying to simplify matters, and doubtless there was a tinge of American vs. British prejudice in it.

  10. Jerry Friedman said,

    May 10, 2011 @ 12:19 pm

    @Dimitri: It's plausible that Latin (or French) was involved in the rarity of split infinitives from the 15th to at least the early 18th century, or in the prescriptions to avoid it that appeared as it became more common in the 19th century. However, I know of no hard evidence for any connection with Latin, nobody saying "You shouldn't do this because it's impossible in Latin" (till the mid 20th century). I'd be interested if anyone knows of one.

    @Morten Jonsson: Glamour may be Scottish but it's not Gaelic. It's a doublet of grammar. You can find several similar versions of its etymology at dictionary.com

    I suspect Scott had clamour particularly in mind. The Follett-Barzun-Wenzberg Modern American Usage suggests that removing the u from the Scottish word glamour is like removing it from dour—but I don't think the cases are analogous.

  11. Dan T. said,

    May 10, 2011 @ 12:42 pm

    To me, most of the British spellings come off with an air of pompous pretentiousness. This seems to be cultivated intentionally by some real estate developers when they name their developments things like "Harbour Centre" (an actual business park in my area).

  12. The Ridger said,

    May 10, 2011 @ 1:11 pm

    Most of Webster's ideas never came to fruition (consider his suggestion that "the words mean, near, speak grieve, zeal, would become meen, neer, speek, greev, zeel", just for one). But regardless of whether any one particular idea was invented by Webster or not, I think it's clear that he had a goal of splitting American off from English. After all, he wrote in 1789 an "An Essay on the Necessity, Advantages, and Practicality of Reforming the Mode of Spelling and of Rendering the Orthography of Words Correspondent to Pronunciation," Dissertations on the English Language: With Notes, Historical and Critical, to Which is Added, by Way of Appendix, an Essay on a Reformed Mode of Spelling, with Dr. Franklin’s Arguments on That Subject" which contained arguments like these:

    4. But a capital advantage of this reform in these states would be, that it would make a difference between the English orthography and the American. This will startle those who have not attended to the subject; but I am confident that such an event is an object of vast political consequence. For,
    The alteration, however small, would encourage the publication of books in our own country. It would render it, in some measure, necessary that all books should be printed in America. The English would never copy our orthography for their own use; and consequently the same impressions of books would not answer for both countries. The inhabitants of the present generation would read the English impressions; but posterity, being taught a different spelling, would prefer the American orthography.
    Besides this, a national language is a band of national union. Every engine should be employed to render the people of this country national; to call their attachments home to their own country; and to inspire them with the pride of national character. However, they may boast of Independence, and the freedom of their government, yet their opinions are not sufficiently independent; an astonishing respect for the arts and literature of their parent country, and a blind imitation of its manners, are still prevalent among the Americans.


    As an independent nation, our honor requires us to have a system of our own, in language as well as government. Great Britain, whose children we are, and whose language we speak, should no longer be our standard; for the taste of her writers is already corrupted, and her language on the decline. But if it were not so, she is at too great a distance to be our model, and to instruct us in the principles of our own tongue.
    It must be considered further, that the English is the common root or stock from which our national language will be derived. All others will gradually waste away-and within a century and a half, North America will be peopled with a hundred millions of men, all speaking the same language. Place this idea in comparison with the present and possible future bounds of the language in Europe–consider the Eastern Continent as inhabited by nations, whose knowledge and intercourse are embarrassed by differences of language; then anticipate the period when the people of one quarter of the world, will be able to associate and converse together like children of the same family.1 Compare this prospect, which is not visionary, with the state of the English language in Europe, almost confined to an Island and to a few millions of people; then let reason and reputation decide, how far America should be dependent on a transatlantic nation, for her standard and improvements in language.

  13. The Ridger said,

    May 10, 2011 @ 1:12 pm

    Sorry, I forgot to source that Webster essay.

  14. GeorgeW said,

    May 10, 2011 @ 1:22 pm

    Vivian Cook ("The English Writing System") says, "The main difference between British and American spelling come down to the reforms of one nineteenth-century man, Noah Webster [. . .] His influential dictionary published in 1828 established several of the distinctly American forms [. . .] Hence he advocated in the preface to the 1828 dictionary:

    -or rather than -our 'honor', 'labor'
    -er rather than -re 'scepter', center'
    -se rather than -ce 'defense', 'offense'

    [. . .] the American spelling -or according to Webster (1928) commenced or received most decided support and authority at the revolution and he cites George Washington's Letters (1795) in support."

  15. J. W. Brewer said,

    May 10, 2011 @ 1:27 pm

    Following up on the interesting quotes about Webster's conscious agenda, the question then becomes which of Webster's proposals in practice actually succeeded and which failed, and why. If the successful subset tended to be the ones which relied on standardization around naturally-occurring extant prior variants (which were at the time being weeded out by standardization around different norms on the other side of the ocean) and the unsuccessful subset tended to be the pure innovations, that would be a lovely parable about the limits of conscious command-and-control in language policy. Down with rationalist constructivism! Down with crypto-Esperantist totalitarianism! Although the question perhaps then becomes why Webster's well-attested loathing for Jacobinism in political matters did not lead him to refrain from proposals which disregarded tradition and organic cultural practice in the orthographic realm.

  16. mollymooly said,

    May 10, 2011 @ 1:40 pm

    @Morten Jonsson: "glamour" is not from Gaelic; it is a Scots by-form of "grammar". The OED's entry on -our lists other words not derived from Latin -or: arbour, armour, demeanour, endeavour, succour, neighbour, harbour.

  17. Sid Smith said,

    May 10, 2011 @ 2:10 pm

    In British naval material (log books, etc) from the Napoleonic period, 'honor' is very common. The OED has Coleridge using this form in 1809-10.

  18. Ray Dillinger said,

    May 10, 2011 @ 2:19 pm

    As an Anglophone, I don't really believe that orthography can be regulated. (a Francophone would probably have a different belief). Certainly there are conventional ways to spell things, but many, many words have many, many acceptable variations. And that was even more true when Webster compiled his dictionary.

    Internal consistency within a document I consider an important point of style; the specific spellings chosen, if within acceptable variations, less so.

    What Webster did was publish a popular dictionary pretty widely, shortly after the idea that words ought to have standard, conventional spellings really took hold. If he wanted America to have an orthography different from England, it was the right thing to do and the right moment to do it. But a dictionarist, or any orthographic reformer, cannot be radical in his proposed reform, or people will reject the work as being inaccurate, having limited utility, or being inconvenient. I don't think his dictionary proposed any great degree of reform as such. What it did was to make choices, usually based on consistency and etymology, between conventional spellings or attested variants of the same word.

    If I were in Webster's place, making a "regular" orthographic change today in hopes of making orthography marginally more closely resemble phonology while making writing marginally more convenient and trying to avoid a proposal too radical for acceptance, I would probably concentrate on reduced vowels in unaccented syllables. Those which are currently spelled with two letters and sound as a single sound, I would respell with one letter. I think maybe a reformer could get away with that. But probably no more than that.

  19. Morten Jonsson said,

    May 10, 2011 @ 2:48 pm


    Thanks for correcting my careless mistake. Next time maybe I'll remember that "Scots" does not necessarily equal "Gaelic."

  20. Spell Me Jeff said,

    May 10, 2011 @ 2:56 pm

    Anglo-American spelling reform will follow Anglo-American adoption of the Euro.

  21. Xmun said,

    May 10, 2011 @ 4:22 pm

    (myl) But the fact that -ize is "Oxford style" — and has been for some time — does tend to undermine the story that -ize was a Webster "reform". (Though I know nothing about whether opinion at Oxford on this question had crystallized before Webster or not…)]

    My understanding is that the Oxford adoption of the -ize ending postdates Webster by a good many years. I was told years ago by Professor I. A. Gordon that a printer at the Press, Horace Hart, insisted on the z spelling (-ize, -ization, etc.) because he considered the letter z to be underused. My edition of "Hart's Rules for Compositors and Readers at the University Press Oxford" is the 38th. It would be good to see what was said about the subject in the first edition of 1914.

  22. Xmun said,

    May 10, 2011 @ 5:19 pm

    A couple of corrections to my last comment. Horace Hart, M.A. was Printer to the University, not merely "a printer at the Press".

    The first edition of his Rules, for internal use only, was printed in 1893. The fifteenth edition, the first for general sale, appeared in 1904. The preface of the 1914 edition is reprinted in the 38th.

  23. Thor Lawrence said,

    May 10, 2011 @ 5:44 pm

    My 1933 7th ed. of the "Authors' & Printers' Dictionary" by F. Howard Collins and probably for the OUP has an entry for -ise that starts "in accordance with the practice of the O.E.D., the following words in this book end in -ise : -ize being generally used ….."
    and then lists the -ise ones.
    There is no entry for -or/-our.

  24. James Kabala said,

    May 10, 2011 @ 11:00 pm

    I didn't keep systematic track of it, but several years ago when I was doing historical research and had to read a large number of Revolutionary/Early Republic pamphlets, I did think I noticed an increase in consistent use of -or and other Websterite forms in the 1830s compared to earlier decades. Perhaps it was just confirmation bias, however.

    Some modern-day unreconstructed Southerners are known to insist on use of -our and -ise as correct, claiming the Websterite forms are Yankee cultural imperialism.

  25. Simon said,

    May 11, 2011 @ 3:24 am

    Could the change in spelling noted in the original query be related to dictionaries becoming more widely available as the technologies of paper production and printing improved greatly at around the same time?

  26. GeorgeW said,

    May 11, 2011 @ 4:45 am

    James Kabala; "Some modern-day unreconstructed Southerners are known to insist on use of -our and -ise as correct, claiming the Websterite forms are Yankee cultural imperialism."

    As a semi-modern, partially reconstructed Southerner, I was unaware of this. But, to the extent this is correct, the motivation could well be the influence of the King James Bible which preserves some of these forms like 'honour' and 'defence.'

  27. klai said,

    May 11, 2011 @ 6:36 am

    @Dimitri – Oh, but it has. Cf. the Latinization (or Latinisation? ;^) ) of words such as debt < det(te), subtle, doubt, indict, receipt, and arctic. In cases like these – and there are many more examples – the silent letter was reintroduced/enforced in their spellings in order to underline the Latin etymology (all of these were borrowed from Old/Middle French).

  28. GeorgeW said,

    May 11, 2011 @ 7:04 am

    James Kabala: Whoops! My comment above (4:45AM) was based on a possible misunderstanding. After thinking about it, you may have been referring to an English Southerner. I am an American Southerner and to us, 'Yankee' is someone from above the Mason-Dixon line.

  29. Andrew (not the same one) said,

    May 11, 2011 @ 9:18 am

    GeorgeW: I'm pretty sure that James Kabala meant American Southerners: I have heard of this phenomenon. It's true that some Brits – not necessarily Southerners – see -ize as American cultural imperialism, not knowing the history of the matter; The Times had to abandon -ize, its traditional form, a few years ago, because people kept complaining about the nasty Americanism. But -or/-our just isn't an issue here.

  30. CNH said,

    May 11, 2011 @ 9:25 am

    I have read a few hundred files in the Public Record Office in Kew, dating to the 1950s and 60s – mainly policy but also technical files. Once or twice I was surprised by an 'ize' where I would nowadays expect an 'ise'.

  31. James Kabala said,

    May 11, 2011 @ 10:55 am

    I did mean American Southerners and meant Yankees as synonymous with Northerners (or more specifically New Englanders), not Americans. I can see how I might have been confusing, but I don't think the issue of reconstructed/unreconstructed ever came up in the Home Counties!

    I was not saying this is a common phenomenon; it seems (as far as I can tell from the Internet, being a Northerner myself) to be limited mostly to members of self-conscious Southern heritage groups. I did once stumble across a polemic against the -ize form by a Southerner who believed he (or maybe it was a she) had clinched his case with the Shakespeare (King Lear) quotation "Thou whoreson zed, thou unnecessary letter!"

  32. Mr Punch said,

    May 11, 2011 @ 11:42 am

    Webster was a prescriptivist and a spelling reformer, at a time when spelling was by no means standardized. The issue is not so much how his prescriptions aligned with contemporary practice among the general public as how they diverged from those of Samuel Johnson and other English authorities.

  33. John Cowan said,

    May 11, 2011 @ 12:32 pm

    "Yankee" refers to the subset of Vermonters who eat pie for breakfast. Any further extensions are mere sloppiness and totally unwarranted.

  34. J. W. Brewer said,

    May 11, 2011 @ 2:11 pm

    An interesting tidbit seen on wikipedia while looking for something else: 'The Honorable Society of King's Inns (HSKI), is the institution which controls the entry of barristers-at-law into the justice system of the Republic of Ireland. The full title retains the historical spelling variant "honorable" in preference to the contemporary Irish/British "honourable."'

  35. Chris Holdaway said,

    May 11, 2011 @ 2:41 pm

    I have read before that the modern accent and phonology of English in the United States, is apparently far closer to the EModE of say, the Shakespearian era in England.

    Couldn't say where I got that information, so it could just be something I heard, and maybe another piece of pop-linguistics that gets touted around.

  36. Jason L. said,

    May 11, 2011 @ 3:49 pm

    mollymooly: The OED's entry on -our lists other words not derived from Latin -or: arbour, armour, demeanour, endeavour, succour, neighbour, harbour.

    "Arbour" doesn't derive from Latin "arbor"?

    (It's interesting to note that, like "glamour", the last two items in the list, "neighbour" and "harbour", don't even derive from Latin at all!)

  37. Dw said,

    May 11, 2011 @ 6:23 pm

    @Chris Holdaway

    I have read before that the modern accent and phonology of English in the United States, is apparently far closer to the EModE of say, the Shakespearian era in England.

    far closer than what? British English?

    This is something of a myth. Both British and American accents of English have diverged away from Shakespearean English in many ways, some shared and some distinct.

    For example, most English people vocalize historical nonprevocalic /r/s and many (probabably most nowadays) also vocalize historical nonprevocalic /l/s.

    On the other hand, most English people preserve some vowel distinctions (e.g. "father"-"bother", "cot"-"caught", "Mary"-"marry"-"merry") that tend to be lost in many American speakers.

    If you are looking for the most conservative accent of English today, you should probably look at southern Irish English.

  38. maidhc said,

    May 11, 2011 @ 7:40 pm

    There's also the Labor Party in Australia. Wikipedia says it was King O'Malley who removed the 'u'.

  39. Debbie said,

    May 12, 2011 @ 3:49 pm

    Could some of the differences have resulted from public education in the US, pushed for by Webster, although not fully implemented for fifty years following independence. And could further divergence have resulted from colonization of India by the French, thus influencing spelling of words?

  40. Chris Holdaway said,

    May 12, 2011 @ 9:37 pm


    Yes I had always assumed that statement was something of a pop-linguistic one. Never investigated it, as diachronic studies, particularly with respect to phonology and pronounciation, are somewhat outside my domain.

    Thanks for the clarification, next time someone brings it up I can slap them with the reality.

  41. This Week’s Language Blog Round-Up | Wordnik ~ all the words said,

    May 13, 2011 @ 6:25 am

    […] or not “mispronunciation constitutes a genuine neologism” (short answer: yes). This sparked another debate over at the Language Log: what is the true origin of American spelling? What caused the change from […]

  42. Dw said,

    May 13, 2011 @ 3:44 pm


    And could further divergence have resulted from colonization of India by the French, thus influencing spelling of words

    I'm pretty confident answering "No" to that question. French influence on English is much more easily attributable to the geographical proximity of France to England, together with the cultural prestige in which the French language was (and to some extent still is) held there.

  43. Warsaw Will said,

    May 15, 2011 @ 1:57 pm

    It's a pity you stopped your Ngram at 1900, as the trend for many verbs starts reversing back to -ize uncannily around 1925, which would be a suitable subject for research in itself. I tried with revolutionise, organise, mechanise, humanise and economise. (I'm an unashamed -iser.)

RSS feed for comments on this post