If you look at texts from a few hundred years ago, they’re almost illegible, what with all the superfluous e’s and y’s running about, the long-S’s (  ſ  ), and so on. Texts from 100 and 120 years ago seem only to have cleared up some punctuation issues (in an original Sherlock Holmes story you’ll find rôle and coöperate) and regularized foreign words (Esquimaux, Corea — I swear I do read things other than Conan Doyle, but I’m thinking of the best example from him). Novels from the 1950s and and 1960s have very different styles than more modern books, but the spelling seems to be identical.

I have two contradictory theories:

  1. Spelling drift is constant, but slow, so you need centuries for really noticeable change to occur.
  2. Spelling drift has been slowed or halted by modern proof-reading and by technology.

If anyone has any evidence for either theory (or a third), I would be glad to hear it.

  • 2
    But "drifting" is so negative. Say: "Spelling is still improving," or "Spelling is still simplifying."
    – GEdgar
    Jul 5, 2011 at 19:26
  • see also this question
    – Louis Rhys
    Nov 9, 2011 at 7:59
  • 1
    @GEdgar, but how can we say that it is 'improving'? Are we to say that the current spelling of words is much better than their counterparts, say, 200 years ago? Are we to say that our current spelling and grammar is way better than Shakespeare or Chaucer?
    – gelolopez
    Nov 3, 2013 at 2:03
  • Long-S is a matter of glyphs rather than spelling; while there were conventions about where they would or would not be used (though personal preferences led some to disregard them in both directions), long-S was always still S, just as closed and open a are always still a.
    – Jon Hanna
    Jan 27, 2014 at 11:24

5 Answers 5


One of these days I'm going to write a long blog about this. There are many reasons for the differences in spelling and why it is changing and will keep changing.

One reason that yu find so many different spellings in Middle English (ME) is that English, as written tung, had, for all practical purposes, ceased to exist shortly after the Norman-French (NF) Takeover in 1066. When English did start to be written again c1150 - c1175, there was little to nothing to go by. The NF orthography was very different than the Anglo-Saxon/Old English orthography. Often yu will find the NF "ou" substituted for the OE "u". Thus wund became wound (the injury). Thu (þu) became thou (and was said as thu, not thow).

The NF had a rule of not benoting a 'u' before an 'n' or 'm', thus sum, cum, munk, tunge, asf became some, come, monk, tongue, asf. They also added 'ue' where not needed; for byspel: prologue and catalogue. The didn't like a word ending in a 'v' so there are words like give and infinitive whereas the silent e normally indicates that the preceding vowel is long, it doesn't in these cases.

Other French-Latin loving "reformers" actually made things worse. The added the silent b to dout to make doubt to link it to its Latin roots but kept the Old French spelling of colour instead of the Latin color. They added the silent s to iland to make island (not even a Latin word!). They did the same thing to the French ile to make isle.

There were and are regional and dialectal pronunciations that led (and lead) to different spellings. There was, and often still isn't, set orthography. The letter 'y' can be either be a long ī or short ĭ. So in old spellings, yu might see thinking or thynkyng ... like or lyk.

There are pedants who are resistant to even the simplest and very logical changes. There is no justification for although, though, enough, and through. The 'ough' cluster is confusing and unnecessary. Even tho the alt spellings of altho, tho, and thru hav been recommended by various reform panels and hav been around for years (and are preferred by some organizations like the US Army), pedants insist that they are not acceptable in formal writings. Why? On what grounds?

(BTW, I used altho, tho, thru all thru undergraduate and graduate work, the military, and in corporate writings with nary a problem.)

So spelling will keep on changing whether led by free-spellers like me, or by accident but not without vitriolic sniping by pedants.

BTW, if yu think that my changes in spelling hurt yur eyes, try this (it's all English):

All menschli bȋings ar born frie and elyk in wurđinis and ryhts. Đei ar gifted wiđ riedschip and inwit and schuld behæv tewards òan anuđer in a mûd ov bruđerhud.

  • 3
    Your changes do not hurt my eyes. It is “ough” that hurts my eyes :-)
    – marcus
    Feb 2, 2012 at 4:54
  • I’d’ve thought that inwit (“conscientia bona” per A.R.) would’ve worked for both reason and conscience, although perhaps better as the latter, considering its overtones of self-awareness. I’m wondering about your spelling of riedschip though. Why not redeschip or even raedescip? Why’d you use ie for ae or plain e, considering how OE rǽd had already rede by Middle English become? Raedelike is wise or sensible, and rede for reason or counsel is still sometimes found: “Yet do not cast all hope away. Tomorrow is unknown. Rede oft is found at the rising of the Sun.”
    – tchrist
    Nov 3, 2013 at 20:07
  • Hm, a rede-ȝive seems to have been “One who gives rede or counsel; a counsellor, adviser.” Life was simpler then. And language.
    – tchrist
    Nov 3, 2013 at 20:16

"Spelling drift is constant, but slow, so you need centuries for really noticeable change to occur."

It slowed immensely with the invention of printing as standardised text became common. But has probably increased recently with mixtures of British and American English being used in the same text.

I think a more obvious factor is the simplification and general reduction of punctuation in written English, especially online.

"Spelling drift has be slowed or halted by modern proof-reading and by technology."

The same technology has produced a lot of informal written English from a much larger section of the world's population. Apart from the number of non-native speakers writing blogs, emails and tweets in English there is a whole new world of new words being produced.

The new technology has probably been the death of few remaining diacritics, nobody is going to bother going through the menus to add a circumflex to "crepe" on a cell phone.

  • That seems to be true, although adding the circumflex (not an accent) to crêpe is much easier on an iPhone (just hold the key down for a second and select from the menu) than it is on a computer keyboard, which entails Googling "e circumflex mac keyboard" ... Jul 5, 2011 at 17:21
  • It's harder on unix because dcs-media.com/Archive/humor-real-programmers-dont-eat-quiche-AI or crêpe
    – mgb
    Jul 5, 2011 at 18:16
  • There is one more logical possibility, "Spelling drift has sped up with more visibility of non-transient, public repositories of text in non-traditional communities". Blogs, twitter, web pages. I personally think your option 2 (spelling change is slowing) is most likely overall (we are all reading more and so have more view of spellings and so we're more likely to learn the majority spelling (which is presumably correct)), but maybe some smaller communities may drift faster.
  • One presumably could test these hypotheses more objectively by doing the experiment of counting average spelling errors...oops...differences in a all calendar years against a single given spelling dictionary. The corpus of data would have to be controlled (that is, not start letting in tweets into a book corpus).

Spelling is still drifting.

Consider the doubled consonants in the words, "traveller" and "worshipper," for instance. These were the preferred forms in the early part of the last century. Then the forms with only a single "l" and "p," respectively, became variants.

Then, those forms became primary with the doubled forms now given as variants. Eventually, they'll fall off altogether.

Recently, we saw the OED list "drive-thru," validating the truncation of "through" at least in this instance. Over time, advanced languages tend to simplify, and we're witnessing that happen right before our eyes in these and in many other cases.


I daresay that your second theory, that "[s]pelling drift has be slowed or halted by modern proof-reading and by technology," is truer if by "drift" you mean that it is deviating from how words are pronounced. If you mean spelling changes still occur despite the fact that we have better ways to keep it in order, then I'd say that it is indeed still changing.

The Raven brought up the example of "drive-thru" entering the OED. This is a case of the simplification that he says is happening actually occurring. With the emergence of the Internet and electronic communication/texting, people tend to opt for spellings that make more sense and that will take us less time to punch in with those microscopic buttons cell phones had and occasionally still have. There is little doubt that it takes less time to spell right as "rite," or night as "nite," or though as "tho," especially if you're typing with a number pad that requires you to hit a certain button several times to arrive at the desired letter. Superfluous letters are being eradicated in favor of more logical spellings, and taking a trip to any online forum, or even to Facebook, will show you this.

PS - Didn't notice this post was over 2 years old, lol.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.