R.I.P. John Holm (1943-2015)

« previous post | next post »

Today's New York Times includes an obituary for the pioneering creolist John Holm, with some remembrances from our own Sally Thomason.

John Holm, a linguist who helped bring the study of creole and pidgin languages into the scholarly mainstream, died on Dec. 28 in Azeitão, Portugal. He was 72.

The cause was prostate cancer, his husband, Michael Pye, said.

While hitchhiking through Mexico and Central America as a teenager, Mr. Holm heard black Nicaraguans along the Caribbean coast speaking a non-Spanish language that seemed oddly familiar. They called it “pirate English,” a reference to its probable origin as a pidgin spoken on pirate and British Navy ships.

Although Mr. Holm could barely understand what he was hearing, it planted the seed for what would become his life’s work: the study of creole and pidgin languages spoken by millions of people around the world, especially the English-derived creoles of the Caribbean.

A pidgin is a reduced language used by groups with no language in common who need to communicate for trade or other purposes. A creole, by contrast, is a natural language developed from a mixture of different languages, like Haitian Creole, which is based on 18th-century French but absorbed elements of Portuguese, Spanish and West African languages. Semi-creole languages, which Mr. Holm also studied — Afrikaans is an example — share even more traits with their vocabulary-source languages.

After compiling the first dictionary of Bahamian English, Mr. Holm produced a landmark study, the two-volume “Pidgins and Creoles,” which traced the socio-historical evolution of pidgins and creoles, explained their structures and described more than 100 varieties.

“Pidgin and creole studies had generally been dismissed, largely because creole languages in particular were thought to be spoken almost exclusively by poor people of color and were considered to be bastardized versions of the European languages that contributed their vocabularies,” said Sarah Thomason, a linguist at the University of Michigan and an associate editor of The Journal of Historical Linguistics.

“By publishing careful empirical studies of creole and semi-creole language structures, and by publicizing these languages in ‘Pidgins and Creoles,’ John helped create one of the most exciting subfields of linguistics,” she added.

You can read the full obituary here, and see Holm's CV for more of his academic accomplishments.

Update: An obituary for Holm also ran in The Bahamas Tribune.


  1. sally thomason said,

    January 4, 2016 @ 6:14 pm

    Just in case it isn't totally clear from the obituary: creoles are NOT spoken almost exclusively by poor people of color and are NOT bastardized versions of their vocabulary-source languages. But that was the perception, even among linguists, all those decades ago when John Holm helped to change the perception dramatically. Now, by contrast and in part due to John's scholarship, creoles (and, to a somewhat lesser extent, pidgins) are regarded by most? all? linguists as fascinating languages in their own right, languages whose genesis (or creation) processes provide uniquely valuable insights into the nature of human language itself — its history and its structure, and the creative behavior of people in language contact situations. John's impact on creole studies, though, was only in part due to his scholarly achievements: he was a wonderfully kind and gentle man, beloved I do believe by everyone who knew him, and he helped to soften the often abrasive controversies that afflict pidgin and creole studies. His encouragement of younger scholars in particular is legendary.

  2. mike said,

    January 4, 2016 @ 9:09 pm

    >creoles are NOT spoken almost exclusively by poor people of color and are NOT bastardized versions of their vocabulary-source languages

    There are people who will maintain that English is a creole, right?

    Anyway, thanks for the writeup–Holm sounds like he was a lovely person.

  3. Kaleberg said,

    January 5, 2016 @ 12:08 am

    I wonder how many other modern languages besides English may be derived from creoles. One would imagine that creoles would have a lot of survival power since they usually develop where two or more other languages meet and are often associated with cultural change.

  4. John Roth said,

    January 5, 2016 @ 1:05 am

    Definitions of "pidgin" and "creole" have become confused recently. If you take the original definitions due to Holm and Bickerton, etc, then a pidgin is a language of convenience that's formed by people who have neither a common language nor another language that they can learn well. A creole is the next generation result of a pidgin, and evolves from that base in the usual manner of any other natural language.

    This definition has the advantage that it's historically attested in numerous instances. Using that definition, the grammar of early generation creoles has intriguing similarities which don't seem to derive from the "official" languages around them.

    Arguments that English went through a creole stage depend on a different definition of creole than that above.

    The Wikipedia article isn't entirely awful as long as you keep the definitional problems in mind.

  5. Coby Lubliner said,

    January 5, 2016 @ 10:58 am

    It can also be argued that all Romance languages are creoles, derived from the pidgin Latin spoken as the means of communication by the many Celtic and other tribes of the Roman Empire.
    The same could be said of modern Hebrew, as described by Amos Oz: the various Jewish communities living in Jerusalem in the 19th century, speaking Ladino, Yiddish, Persian, Arabic etc. could communicate with one another only in the rudimentary Hebrew that they had learned in school, out of which developed a common language that was finally codified by Ben Yehuda.

  6. DWalker said,

    January 5, 2016 @ 1:04 pm

    The terms "pidgin" and "creole" might be undefinable.

    "It can also be argued that all Romance languages are creoles, derived from the pidgin Latin spoken as the means of communication by the many Celtic and other tribes of the Roman Empire."

    You could call Latin a creole too, right? And Arabic, and Hebrew?

  7. James Parkin said,

    January 5, 2016 @ 1:19 pm

    I don't think it's right to regard Romance languages as creoles. Though they simplified the Latin gender system and lost most of the classical nominal inflection, they retained, even through changes, a very complex verbal morphology. Is this something one would see in a pidgin/creole development?

  8. J. W. Brewer said,

    January 5, 2016 @ 1:46 pm

    Insofar as a majority of humans alive on the planet are "poor people of color" (I guess with the percentage varying depending on how you define poverty and "color"), there are lots and lots of languages spoken primarily-to-exclusively by poor people of color, many of which (e.g. the indigenous languages of the Americas) had long been thought respectable subjects of study for pale and prosperous academics. So I think an account for why creoles were once thought unworthy of academic study needs to be a bit more refined. If anything, one could hypothesize a certain prosperous-white-liberal-academic valorization of the "noble savage" in his pristine original setting, such that the languages of West Africa might be thought quite interesting, but the new language varieties that the descendants of the speakers of those languages subsequently created for themselves in the Caribbean etc. would instead be thought merely debased versions of European languages that resulted from the debasement and humiliation of their speakers at the hands of their European oppressors and thus rather embarrassing for all concerned. It was the oppressors' fault that they don't speak their ancestral languages and likewise the oppressors' fault that they haven't been given a proper education such that their L1 is a standard/prestige variety of the relevant-in-context European language. Neither of those moral judgments is necessarily wrong as a moral judgment, but (even bracketing the issue of the appropriateness of moral judgment in the scientific study of language) they ignore the perspectives both that: a) pretty much all varieties of human language are inherently fascinating (ok, maybe that's a matter of taste, but if you don't think so, what are you doing in the linguistics business?); and b) creoles can be easily understood (if you feel the need for an uplifting moral narrative in your Sprachwissenschaft) as a testimony to how innovative humans are often capable of being under objectively lousy social circumstances.

  9. maidhc said,

    January 5, 2016 @ 8:15 pm

    If you go back in history, any minority language spoken mostly by poor people and with little written literature was considered a debased jargon unworthy of serious academic study: Welsh, Breton, Gaelic, Finnish, Czech, Roma, etc. It wasn't until the 19th century that some people began to take these languages more seriously, in many cases starting by writing down the oral literature of those languages.

    In addition, creoles suffer from the "it's not even a proper language" prejudice.

    There were some people taking an interest in creoles earlier on, such as Zora Neale Hurston, who worked in Jamaica, Haiti and Honduras, but not really a detailed linguistic study.

  10. Jason said,

    January 5, 2016 @ 10:59 pm

    The terms "pidgin" and "creole" are probably scientifically meaningless. They are terms of art in Bickertonian Universal Grammar theory, where "pidgin" is an ad-hoc improvised language and a "creole" is a new Chomsky-conformant language with native speakers in conformance with the bioprogram, but Bickerton's account has come in for some serious scrutiny — for example, in "Substratal Influence in Saramaccan Serial Verb Constructions" McWhorter shows that the serial verb constructions in Saramaccan, at least, can be explained by substrate influence and not some bioprogram universality, as alleged by Bickerton. Suzanne Romaine's work demonstrates pretty clearly that most of Tok Pisin's key grammar was settled before there were any significant numbers of native speakers whatsoever — Tok Pisin and related Bislama were used primarily by adult "blackbirded" laborers and later as an interlanguage between white colonalists and Melanesians, nearly always acquired by both parties as adults. Likewise, Gil'Ad Zuckerman account of the development of what he calls "Israeli" in no way follows any neat bioprogram hypothesis and is more likely to be simply a blending of the various L1 languages spoken by the dominant Zionist groups at the time of settlement in a way that is neither "creole" nor "mixed language" by Bickertonian definition.
    As far as the argument that "Creoles" were spoken mostly by poor people and hence weren't considered serious objects of study, as J. W. Brewer has pointed out, linguists studied the indigenous languages of the Americas (among others) with intense interest — indeed, the study of non Indo-European languages by "poor people" could be considered the birth of the discipline, since the existing theoretical frameworks in philology were tested to breaking point by these exotic grammars. No, maidhc, we are not talking about the pre-linguistics era when all languages were seen as more or less failed versions of Latin and Greek, but the 19th and 20th century up to the 1950s.
    So the notion of cultural obliviousness due to these being the languages of "poor people" makes a neat liberal fable, but is nothing more than that. Far more likely is the hypothesis that people in all cultures have universal cultural aversion towards the "incorrect" use of their language characteristic of creolized versions of it. Thus speakers of Swahili react with visceral negativity towards Sheng, the urban creolized version of Swahili in Kenya; native speakers of Welsh rail against the "incorrect" (Anglicised) Welsh in official documents and signs; Israeli/Modern Hebrew speakers rail against barbarous constructions (see http://www.zuckermann.org/pdf/abba.pdf ). While sometimes this relationship of stigmatization is between "richer" and "poorer" speakers, it doesn't always align that way: sometimes the preferred feature is characteristic of "poor" speakers, as in one of Zuckermann's examples, or this attitude can be internalized; Jamacian patois speakers often view their own language as mere "broken" English. It can be reciprocal between feuding group as evidenced by the heat produced by this language-shaming incident: http://www.pri.org/stories/2012-02-21/growing-rift-between-hong-kong-and-mainland-china ).
    I think the explanation is that creoles (and mixed languages, and various sociolects) invoke the "uncanny valley" effect to speakers of the lexifier language; they resemble the lexifier sufficiently that any variations from it are jarring and sound like ignorant barbarisms, "mere jargon", rather than being a rich language with an alternate grammar. Indeed, they resemble the incorrect language produced by L2 learners of that language, and can be seen as regularization of such "incorrect" L2 learning.

  11. DWalker said,

    January 6, 2016 @ 1:14 pm

    My point was, which languages would NOT be considered creoles? And why not?

    Did each language develop from some other language, back to the first speaking humans? Or from the events at Babel? :-)

RSS feed for comments on this post