Expresso, the Computational Sequel.

We discussed the espresso/expresso thing briefly in 2014 (though most of the thread is on weird pronunciations); now Vitaliy Kaurov (of Wolfram Science and Innovation Initiatives) has a much deeper dive that quickly gets too technical for me, Finding X in Espresso: Adventures in Computational Lexicology. But I’m sure some of my readers will happily go into the details; me, I just enjoy the pretty graphs, and I will share the largely comprehensible conclusion:

The following factors affirm why expresso should be allowed as a valid alternative spelling.

Espresso/expresso falls close to the median usage frequencies of 2,693 official alternative spellings with Levenshtein EditDistance equal to 1
• The frequency of espresso/expresso usage as whole pair is above the median, so it is more likely to be found in published corpora than half of the examined dataset
• Many nearest neighbors of espresso/expresso in the frequency space belong to a basic vocabulary of the most frequent everyday usage
• The history of espresso/expresso usage in English corpora shows simultaneous growth for both spellings, and by temporal pattern is reminiscent of many other official alternative spellings
• The uniqueness of the sx mutation in the espresso/expresso pair is typical, as numerous other rare and unique mutations are officially endorsed by dictionaries

So all in all, it is ultimately up to you how to interpret this analysis or spell the name of the delightful Italian drink. But if you are a wisenheimer type, you might consider being a tinge more open-minded. The origin of words, as with the origin of species, has its dark corners, and due to inevitable and unpredictable language evolution, one day your remote descendants might frown on the choice of s in espresso.

Thanks, Kobi!


  1. Graphics are really cool and conclusion may very well be valid. But I am not convinced. Most of the peer group of espresso/expresso in the article is of the AmE/BrE type and for the words endings, not the first consonant. And most of those spelling differences are just spelling conventions, not reflecting differences in pronunciation, because of the clear presence of absence of the latter. I mean, I am all for non-linguists doing linguistic research and completely ignoring basic linguistic stuff clear even to a naked eye. That gives me a chance (I couldn’t produce such cool graphics though, that seems to be the conundrum, either one have to be an expert or make cool graphs), but for the specific question at hand, I would simply defer to M-W.

  2. As Kaurov clearly understands (see the first two sentences of his article) the question is not spelling. It’s pronunciation as well. There are two different words for the same drink: espresso, the Italian loan-word, and expresso, a purely English-language invention.

  3. Yes, he understands, but he completely ignores it in his numerical analysis. Also, they are not quite different words, they are variants of the same word like ask and aks, et cetera and exetera, else and eltze. In all these other examples, only standard spelling is correct, though pronunciation varies. If someone wants to make a case that those are not good analogies and expresso is an eggcorn or folk etymology, I am ready to listen, but to thrust it into the swarm of variant spellings is not a good idea.

  4. wisenheimer

    A mock loan-word “weisenheimer” further Americanized. Harvey Winstin didn’t take that route.

  5. Yes the graphics are cool. But I can’t see where he’s drawing his data from. It should be a well-curated corpus of actual usages.

    I have a nervous feeling he’s counting entries in dictionaries: “The starting point is a dictionary that is represented in the Wolfram Language by WordList and contains 84,923 definitions:”

    Whereas some of the plots of frequency of espresso vs expresso show espresso orders-of-magnitude more frequent. (Beware the logarithmic scales!)

    Some of his other ‘variant’ spellings are U.S. vs Br.E. usages (especially z vs s) or I’d have no scruple in saying just plain wrong: programming vs programing — confusion with programme (as in T.V.) vs program (a computer).

    Not that I’m disagreeing with Kaurov’s comments. (I’m no spelling/pronounciation peever.) But I suspect he could have made them anyway without looking at any data. The whole thing seems to be publicity for Wolfram, by the way — including Wolfram’s (proprietary?) “typical published English text”.

  6. As an undergraduate fifty years ago, I had to endure a “programed text” of logic–i.e. the subject was “taught” behavioristically via a set of propositions followed by fill-in-the-blanks questions. I considered the spelling in the subtitle to be of a piece with the authors’ whole approach to the complexity of reality.

  7. Eli Nelson says:

    programming vs programing — confusion with programme (as in T.V.) vs program (a computer)

    @AntC: Huh, what do you mean? And which do you think is “just plain wrong”? Certainly “programme” as a verb would have to have the gerund/present participle “programming”, but I think “programme” is rarely used as a verb. For the verb “program”, I don’t know of any clear way to establish that “programming” is incorrect: final-consonant doubling occurs less consistently after unstressed syllables than after stressed syllables, but it is known to occur, especially when the vowel is unreduced: compare worshipping, kidnapping, formatted. And “programming” as an inflected form of “program” certainly has plenty of usage—much more than “programing” in fact.

  8. “Expresso” just doesn’t sound right to my ears. There are also people who say “excape”. Could we say that that is correct? Where do linguists and lexicographers draw the line? What can be classified as wrong anymore?

    “Espresso” should be written and pronounced as such because it originated in Italy, and that’s how it is spelled and pronounced in Italian.

  9. So you think all loan words should be written and pronounced as they are in the language of origin? Do you think we should say and write Moskva instead of Moscow, or do you insist on Москва for the written form?

  10. Moskva, how much is in this sound…
    Its “m” and “o” and “s” and “kva”.

  11. Trond Engen says:

    Where do linguists and lexicographers draw the line?

    They don’t. But depending on the type of dictionary, they may say something about register, or connotations, or how the form came to be.

    What can be classified as wrong anymore?

    Lexicographically? Nothing. It’s not the lexicographer’s job to do that. It’s whoever decides that a certain lexicographical collection is the standardised form of a language. Big and important caveats aside, having a standard form is a Good Thing. The problems come when we believe that this standard language is in some sense refined, or cultivated, or generally better than any other form, rather than an arbitrarily chosen temporary convention. And it’s an easy thing to believe. We spend our formative years being punished for every deviation and celebrated as masters of good language when we adhere. But the deviations we hear are the seeds of next year’s grammar. Language moves on, as it did before us, with random permutations and new systems arising, and the lexicographers take new notes, which find their way into a new edition of the dictionary. And a new generation will cling to it.

  12. Well said.

  13. @Eli sorry to be unclear.

    It’s “programing” I think is just plain wrong. (So I’m agreeing with you?)

    “Programme” is rarely used as a verb. I disagree with rarely; it’s less common only because few of us plan T.V. schedules or conferences/timetables, or Public Works — a Programme of Works. (I’m Br.E.)

    “Program” as a noun or verb I use only for computers.

    I was suggesting that “programing” is a mal-formation from adding -ing to “program”. Without the doubled consonant, it would be pronouced with ‘long a’.

  14. marie-lucie says:

    More coffee: The French equivalent of Italian espresso is un express.

    The French word is also used for a train which neither stops at every rural village nor zips through the country as fast as possible, only stopping at a few major cities.

  15. The Taxing Etymology of Ask

    An excellent read, thanks for posting it — I didn’t know about the “ash” form!

  16. The café by my house in Tel Aviv when I was a kid called milkshake “expresso”. I don’t know if they had coffee-type espresso as well.

Speak Your Mind