Arrant Pedantry

By

Cognates, False and Otherwise

A few months ago, I was editing some online German courses, and I came across one of my biggest peeves in discussions of language: false cognates that aren’t.

If you’ve ever studied a foreign language, you’ve probably learned about false cognates at some point. According to most language teachers and even many language textbooks, false cognates are words that look like they should mean the same thing as their supposed English counterparts but don’t. But cognates don’t necessarily look the same or mean the same thing, and words that look the same and mean the same thing aren’t necessarily cognates.

In linguistics, cognate is a technical term meaning that words are etymologically related—that is, they have a common origin. The English one, two, three, German eins, zwei, drei, French un, deux, trois, and Welsh un, dau, tri are all cognate—they and words for one, two, three in many other language all trace back to the Proto-Indo-European (PIE) *oino, *dwo, *trei.

These sets are all pretty obvious, but not all cognates are. For example, the English four, five, German vier, fünf, French quatre, cinq, and Welsh pedwar, pump. The English and German are still obviously related, but the others less so. Fünf and pump are actually pretty close, but it seems a pretty long way from four and vier to pedwar, and an even longer way from them to quatre and cinq.

And yet these words all go back to the PIE *kwetwer and *penkwe. Though the modern-day forms aren’t as obviously related, linguists can nevertheless establish their relationships by tracing the them back through a series of sound changes to their conjectured historical forms.

And not all cognates share meaning. The English yoke, for instance, is related to the Latin jugular, the Greek zeugma, and the Hindi yoga, along with join, joust, conjugate, and many others. These words all trace back to the PIE *yeug ‘join’, and that sense can still be seen in some of its modern descendants, but if you’re learning Hindi, you can’t rely on the word yoke to tell you what yoga means.

Which brings us back to the German course that I was editing. Cognates are often presented as a way to learn vocabulary quickly, because the form and meaning are often similar enough to the form and meaning of the English word to make them easy to remember. But cognates often vary wildly in form (like four, quatre, and pedwar) and in meaning (like yoke, jugular, zeugma, and yoga). And many of the words presented as cognates are in fact not cognates but merely borrowings. Strictly speaking, cognates are words that have a common origin—that is, they were inherited from an ancestral language, just as the cognates above all descend from Proto-Indo-European. Cognates are like cousins—they may belong to different families, but they all trace back to a common ancestor.

But if cognates are like cousins, then borrowings are like clones, where a copy of word is taken directly from one language to another. Most of the cognates that I learned in French class years ago are actually borrowings. The English and French forms may look a little different now, but the resemblance is unmistakable. Many of the cognates in the German course I was editing were also borrowings, and in many cases they were words that were borrowed into both German and English from French:

bank
drama
form
gold
hand
jaguar
kredit
land
name
park
problem
sand
tempo
wind
zoo

Of these, only gold, hand, land, sand, and wind are actually cognates. Maybe it’s nitpicking to point out that the English jaguar and the German Jaguar aren’t cognates but borrowings from Portuguese. For a language learner, the important thing is that these words are similar in both languages, making them easy to learn.

But it’s the list of supposed false cognates that really irks me:

bad/bath
billion/trillion
karton/cardboard box
chef/boss
gift/venom
handy/cellphone
mode/fashion
peperoni/chili pepper
pickel/zit
rock/skirt
wand/wall
beamer/video projector
argument/proof, reasons

The German word is on the left and the English word on the right. Once again, many of these words are borrowings, mostly from French and Latin. All of these borrowings are clearly related, though their senses may have developed in different directions. For example, chef generally means “boss” in French, but it acquired its specialized sense in English from the longer phrase chef de cuisine, “head of the kitchen”. The earlier borrowing chief still maintains the sense of “head” or “boss”.

(It’s interesting that billion and trillion are on the list, since this isn’t necessarily an English/German difference—it also used to be an American/British difference, but the UK has adopted the same system as the US. Some languages use billion to mean a thousand million, while other languages use it to mean a million million. There’s a whole Wikipedia article on it.)

But some of these words really are cognate with English words—they just don’t necessarily look like it. Bad, for example, is cognate with the English bath. You just need to know that the English sounds spelled as <th>—like the /θ/ in thin or the /ð/ in then—generally became /d/ in German.

And, surprisingly, the German Gift, “poison”, is indeed cognate with the English gift. Gift is derived from the word give, and it means “something given”. The German word is essentially just a highly narrowed sense of the word: poison is something you give someone. (Well, hopefully not something you give someone.)

On a related note, that most notorious of alleged false cognates, the Spanish embarazado, really is related to the English embarrassed. They both trace back to an earlier word meaning “to put someone in an awkward or difficult situation”.

Rather than call these words false cognates, it would be more accurate to call them
false friends. This term is broad enough to encompass both words that are unrelated and words that are borrowings or cognates but that have different senses.

This isn’t to say that cognates aren’t useful in learning a language, of course, but sometimes it takes a little effort to see the connections. For example, when I learned German, one of my professors gave us a handout of some common English–German sound correlations, like the th ~ d connection above. For example, if you know that the English /p/ often corresponds to a German /f/ and that the English sound spelled <ea> often corresponds to the German /au/, then the relation between leap and laufen “to run” becomes clearer.

Or if you know that the English sound spelled <ch> often corresponds with the German /k/ or that the English /p/ often corresponds with the German /f/, then the relation between cheap and kaufen “to buy” becomes a little clearer. (Incidentally, this means that the English surname Chapman is cognate with the German Kaufmann.) And knowing that the English <y> sometimes corresponds to the German /g/ might help you see the relationship between the verb yearn and the German adverb gern “gladly, willingly”.

You don’t have to teach a course in historical linguistics in order to teach a foreign language like German, but you’re doing a disservice if you teach that obviously related pairs like Bad and bath aren’t actually related. Rather than teach students that language is random and treacherous, you can teach them to find the patterns that are already there. A little bit of linguistic background can go a long way.

Plus, you know, real etymology is a lot more fun.

Edited to add: In response to this post, Christopher Bergmann (www.isoglosse.de) created this great diagram of helpful cognates, unhelpful or less-helpful cognates, false cognates, and so on:

Click to see the full-sized image.

By

The Drunk Australian Accent Theory

Last week a story started making the rounds claiming that the Australian accent is the result of an “alcoholic slur” from heavy-drinking early settlers. Here’s the story from the Telegraph, which is where I first saw it. The story has already been debunked by David Crystal and others, but it’s still going strong.

The story was first published in the Age by Dean Frenkel, a lecturer in public speaking and communications at Victoria University. Frenkel says that the early settlers frequently got drunk together, and their drunken slur began to be passed down to the rising generations.

Frenkel also says that “the average Australian speaks to just two thirds capacity—with one third of our articulator muscles always sedentary as if lying on the couch”. As evidence, he lists these features of Australian English phonology: “Missing consonants can include missing ‘t’s (Impordant), ‘l’s (Austraya) and ‘s’s (yesh), while many of our vowels are lazily transformed into other vowels, especially ‘a’s to ‘e’s (stending) and ‘i’s (New South Wyles) and ‘i’s to ‘oi’s (noight).”

The first sentence makes it sound as if Frenkel has done extensive phonetic studies on Australians—after all, how else would you know what a person’s articulator muscles are doing?—but the claim is pretty far-fetched. One-third of the average Australian’s articulator muscles are always sedentary? Wouldn’t they be completely atrophied if they were always unused? That sounds less like an epidemic of laziness and more like a national health crisis. But the second sentence makes it clear that Frenkel doesn’t have the first clue when it comes to phonetics and phonology.

There’s no missing consonant in impordant—the [t] sound has simply been transformed into an alveolar flap, [r], which also happens in some parts of the US. This is a process of lenition, in which sounds become more vowel-like, but it doesn’t necessarily correspond to laziness or lax articulator muscles. Austraya does have a missing consonant—or rather, it has a liquid consonant, [l], that has been transformed into the semivowel [j]. This is also an example of lenition, but, again, lenition doesn’t necessarily have anything to do with the force of articulation. Yesh (I presume for yes) involves a slight change in the placement of the tip of the tongue—it moves slightly further back towards the palate—but nothing to do with the force of articulation.

The vowel changes have even less to do with laziness. As David Crystal notes in his debunking, the raising of [æ] to [ε] in standing actually requires more muscular energy to produce, not less. I assume that lowering the diphthong [eɪ] to [æɪ] in Wales would thus take a little bit less energy, but the raising and rounding of [aɪ] to [ɔɪ] would require a little more. In other words, there is no clear pattern of laziness or laxness. Frenkel simply assumes that there’s a standard for which Australians should be aiming and that anything that misses that standard is evidence of laziness, regardless of the actual effort expended.

Even if it were a matter of laziness, the claim that one-third of the articular muscles are always sedentary is absolutely preposterous. There’s no evidence that Frenkel has done any kind of research on the subject; this is just a number pulled from thin air based on his uninformed perceptions of Australian phonetics.

And, again, even if his claims about Australian vocal laxness were true, his claims about the origin of this supposed laxness are still pretty tough to swallow. The early settlers passed on a drunken slur to their children? For that to be even remotely possible, every adult in Australian would have had to be drunk literally all the time, including new mothers. If that were true, Australia would be facing a raging epidemic of fetal alcohol syndrome, not sedentary speech muscles.

As far as I know, there is absolutely zero evidence that Australian settlers were ever that drunk, that constant drunkenness can have an effect on children who aren’t drinking, or that the Australian accent has anything in common with inebriated speech.

When pressed, Frenkel attempts to weasel out of his claims, saying, “I am telling you, it is a theory.” But in his original article, he never claimed that it was a theory; he simply asserted it as fact. And strictly speaking, it isn’t even a theory—at best it’s a hypothesis, because he has clearly done no research to substantiate or verify it.

But all this ridiculousness is just a setup for his real argument, which is that Australians need more training in rhetoric. He says,

If we all received communication training, Australia would become a cleverer country. When rhetoric is presented effectively, it enables content to be communicated in a listener-friendly environment, with well-chosen words spoken at a listenable rate and with balanced volume, fluency, clarity and understandability.

Communication training could certainly be a good thing, but again, there’s a problem—this isn’t rhetoric. Rhetoric is the art of discourse and argumentation; what Frenkel is describing is more like diction or elocution. He’s deploying bad logic and terrible linguistics in service of a completely muddled argument, which is that Australians need to learn to communicate better.

In the end, what really burns me about this story isn’t that Frenkel is peddling a load of tripe but that journalists are so eager to gobble it up. Their ignorance of linguistics is disappointing, but their utter credulousness is completely dismaying. And if that weren’t bad enough, in an effort to present a balanced take on the story, journalists are still giving him credence even when literally every linguist who has commented on it has said that it’s complete garbage.

Huffington Post ran the story with the subhead “It’s a highly controversial theory among other academics”. (They also originally called Frenkel a linguist, but this has been corrected.) But calling Frenkel’s hypothesis “a highly controversial theory among other academics” is like saying that alchemy is a highly controversial theory among chemists or that the flat-earth model is a highly controversial theory among geologists. This isn’t a real controversy, at least not in any meaningful way; it’s one uninformed guy spouting off nonsense and a lot other people calling him on it.

In the end, I think it was Merriam-Webster’s Kory Stamper who had the best response:

aussieaccent

By

Fifty Shades of Bad Grammar Advice

A few weeks ago, the folks at the grammar-checking website Grammarly wrote a piece about supposed grammar mistakes in Fifty Shades of Grey. Despite being a runaway hit, the book has frequently been criticized for its terrible prose, and Grammarly apparently saw an opportunity to fix some of the book’s problems (and probably sell its grammar-checking services along the way).

The first problem, of course, is that most of the errors Grammarly identified have nothing to do with grammar. The second is that most of their edits not only fail to fix the clunky prose but actually make it worse.

Mark Allen already took Grammarly to task in a post on the Copyediting blog, saying that their edits “lack restraint”, that “the list is full of style choices and non-errors”, and that “it fails to make a case for the value of proofreading, and, by association, . . . reflects poorly on the craft of copyediting.” I agreed and thought at the time that nothing more needed to be said.

But then Grammarly decided to go even further. In this infographic, they claim to have found “similar gaffes” in the works of authors ranging from Nicholas Sparks to Shakespeare.

The first edit suggests that Nicholas Sparks needs a comma in the sentence “I am a common man with common thoughts and I’ve led a common life.” It’s true that this is a compound sentence, and such sentences typically require a comma between the two independent clauses. But The Chicago Manual of Style says that the comma can be omitted when the clauses are short and closely related. This isn’t an error so much as a style choice.

Incidentally, Grammarly says that “E. L. James is not the first author to include a comma in her work when a semi-colon would be more appropriate, or vice versa.” But the supposed error here isn’t that James used a comma when she should have used a semicolon; it’s that she didn’t use a comma at all. (Also note that “semicolon” is not spelled with a hyphen and that the comma before “or vice versa” is not necessary.)

Error number 2 is comma misuse (which is somehow different from error number 1, which is also comma misuse). Grammarly says, “Many writers forget to include a comma when one is necessary, or include a comma when it is not necessary.” (By the way, the comma before “or include a comma when it is not necessary” is not necessary.) The supposed offender here is Hemingway, who wrote, “We would be together and have our books and at night be warm in bed together with the windows open and the stars bright.” Grammarly suggests putting a comma after “at night”, but that would be a mistake.

The sentence has a compound predicate with three verb phrases strung together with ands. Hemingway says that “We would (1) be together and (2) have our books and (3) at night be warm in bed together with the windows open and the stars bright.” You don’t need a comma between the parts of a compound predicate, and if you want to set off the phrase “at night”, then you need commas on both sides: “We would be together and have our books and, at night, be warm in bed together with the windows open and the stars bright.” But that destroys the rhythm of the sentence and interferes with Hemingway’s signature style.

Error number 3 is wordiness, and the offender is Edith Wharton, who wrote, “Each time you happen to me all over again.” Grammarly suggests axing “all over”, leaving “Each time you happen to me again”. But this edit doesn’t fix a wordy sentence so much as it kills its emphasis. This is dialogue; shouldn’t dialogue sound like the way people talk?