The Word The Internet Didn’t Know

25 June 2015

Not really.

The word parbunkells got a flurry of press coverage starting a few days ago, such as this piece from Popular Mechanics. Artist Julia Weist rented a billboard in Queens to feature the word, claiming that it was a forgotten seventeenth-century word that did not appear on the internet. Gizmodo declared the word to be “dead to the digital world—and to almost every living person.” Weist was trying to make a point about how information is shared over the internet, telling Gizmodo:

The word has also become a shortcut to a portrait of meaning making and content production on the Internet, both human and non-human, in the sense that you can search for it and see spools of information, reaction, conversation, re-context- ualization and response. In that sense it’s all or nothing, and now that word has been used, the more usage the better.

But she chose a bad example to make her point, and the mainstream media covering the story got a lot wrong. 

The story is kind of an object lesson in not making claims until you’ve done some research or at least spoken to experts in the field. Parbunkells, or parbuckle as it is more usually spelled now, is not a forgotten word. It does have a life of its own on the internet and in meatspace. It’s even in the OED. And if you Google parbunkell, the search engine offers help by asking, “Did you mean parbuckle?” So the word, while rare, wasn’t exactly hiding. Two minutes of poking about would have turned it up.

The OED defines the word as, “a rope, cable, etc., arranged like a sling, used to raise or lower heavy objects vertically.” Popular Mechanics and many of the mainstream outlets that reported the story gave its definition as the vague, “coming together through the binding of two ropes,” proof that writing a good definition is an acquired skill. The word, used mainly in naval and maritime jargon, is attested to as early as 1625, and the early spellings are parbunkel or parbuncle. The word’s origin is uncertain. It may be a borrowing of a Scandinavian term. The par- element is related to pair. The -bunkel is the uncertain bit. In the seventeenth century the spelling parbuckle began to appear, a folk etymology or eggcorn created out of confusion with buckle. The parbuckle spelling quickly became the standard.

The word is first recorded in 1625 in Henry Mainwaring’s Nomenclator Navalis:

A Parbuncle is a rape which is used in ye nature of a paire of Slings.

It also appears in John Smith’s 1627 A Sea Grammar, which is evidently where Wiest found it. Mainwaring’s book exists as a manuscript in the British Library, but has been reprinted at various times over the centuries. Smith’s book is available via Early English Books Online.

The word is still in use. The OED has a citation from as late as 1997 in the Daily Telegraph.

Gizmodo asks, “It’s easy to fall into the trapping of thinking the internet knows everything, but it doesn’t. Oddities like this makes you wonder how much other knowledge is lying on dusty shelves, waiting to be rediscovered.” To which I respond, does Gizmodo know how to use the internet?

Teaching Registers

23 June 2015

The Economist’s Prospero blog has a post on the necessity of teaching different registers of speech. It uses Portuguese as an example, which I can’t speak to, not knowing the language, but the fundamental point the article makes is a good one: “Instead of a rigid right-wrong approach, with the written form always being taught as right, it would be better to teach the idea of register: that certain forms are used in casual speech, other forms in formal speech, others still in writing.”

It’s a good point. Students are smart, and they instinctively know how to switch registers—they do it all the time in their own speech. The only thing that needs to be done is make them aware that they do it. It’s not a difficult concept.

Depicting World Languages

19 June 2015

Language visualization created by Alberto Lucas Lopéz for the South China Morning Post

Language visualization created by Alberto Lucas Lopéz for the South China Morning Post

A neat visualization of the twenty-three most popular languages, depicted proportionally by the number of speakers.

The graphic was created by Alberto Lucas Lopéz for the South China Morning Post.

Data like this is always a bit suspect, but this chart is based on the information at Ethnologue.com, which is generally pretty good. The biggest problem is that it represents only the top twenty-three languages, leaving out the other six thousand or so. It also only captures L1, or first-language users. The total number of English speakers, for example, is much larger. It also fails to capture dialectal differences; for example Chinese is not as unified as the chart makes it out to be. Still, it’s a useful visualization in many ways.

For me, the most surprising thing in the image is the realization that French has relatively few speakers. There are some 87 million additional L2 speakers, but that’s still not a lot compared to the other languages on the chart. I would have guessed that it would have been much higher. Still, in the rankings of all world languages, French is in the top one percent.

Book Review: The Language Myth

14 May 2015

Vyvyan Evans’s The Language Myth is something of a polemic. In the book Evans, a professor of linguistics at Bangor University in the UK, takes on the dominant paradigm of twentieth century linguistics, the universal grammar of Noam Chomsky, especially as popularized by Steven Pinker in books like The Language Instinct. Evans’s book is, to say the least, controversial, and I am not fully qualified to judge its merits. (But this being the internet, I’m going to anyway.)

The book is clearly written, engaging, and accessible to those without formal training in linguistics. So those readers of this website without such training should have no problem understanding Evans’s arguments. But the thing that kept nagging me as I read it was that the view presented is very one sided. I am not certain that Evans has accurately described universal grammar and the arguments behind it. It may be that at times Evans is tilting at something of a straw man or taking a since-retracted or out-of-context statement as definitive of the Chomskyan position. (This is just the impression I got from reading it; it may well be that Evans is spot on.)

Much of Evans’s argument hinges on the meaning of instinct and innate and the difficult task of separating that which is present at birth from that which is learned in infancy and earliest childhood. Unlike Chomsky, Evans argues that humans have a variety of cognitive faculties out of which language emerges. While language relies on specific functions of the brain, it is a social construct that is learned, and not a biological module in the brain that is unique to humans. He bases his conclusions on several sub-fields of linguistics. One is animal studies, the fact that all of the cognitive functions that enable language are found in various animal species—although not to the degree and refinement found in humans, and no known animal species has anything close to the sophistication of human language. Another is the study of how children acquire language—and it is this portion of Evans’s book that I find most compelling in refuting Chomskyan universal grammar. (Notably, it is the least polemical of the sections too.) Evans does a superb job of laying out what we know about language acquisition and how it flies in the face of Chomsky’s theory.

Evans also delves into neo-Whorfian theory and linguistic relativism. It is here that I dispute his argument somewhat, or rather a particular failure in his argument, and do so on firmer ground as I have more familiarity with this sub-field of linguistics. Evans does an excellent job of laying out the experimental evidence for neo-Whorfian relativism, such things as studies in color recognition and how it is effected by one’s native language. His description of the experiments is excellent, except in one respect. He omits any discussion of effect size. Many of the neo-Whorfian relativistic effects that have been shown to exist, while real and statistically significant, are very, very small. For example, a person with a name for a particular shade can recognize it faster than one without, but only about 1/100th of a second faster on average. It’s a measurable effect and relevant to our understanding of the basics of cognition, but it probably has no practical effect on how we live our lives. (Those in the medical field make the distinction between statistically significant and clinically significant, and perhaps linguists should take a tip from them.) To be fair, Evans does not state that there is much of a practical effect, but given how the Whorfian hypothesis has been misconstrued in the popular imagination and the mainstream press, leaving out discussion of effect size is a significant omission.

For those interested in the intersection of language and cognition, Evan’s The Language Myth is something of a must-read. But it should be read with the understanding that it presents just one view in a complex argument.

Evans, Vyvyan. The Language Myth: Why Language is Not an Instinct. Cambridge: Cambridge University Press, 2014.