Login
Helping verbs are curious, AND fascinating
English has a big bagful of auxiliary verbs. You may have learned these as “helping verbs” in elementary and middle school, since they are sometimes described as verbs that “help” the main verb express its meaning. There are even schoolroom songs about them. They are a curious bunch.
The auxiliaries include the modal verbs (can and could, shall and should, will and would, may and might, and must). The verb that follows a modal is in its bare, uninflected form: can go, could go, must go, and so on. There are also a number of semi-modal auxiliary verbs (such as dare, need, ought to, had better, have to, and used to). Some are compound words spelled with a space and several have unusual grammatical properties as well, such as being resistant to contraction or inversion. And in parts of the English-speaking world, modals can double up, yielding expressions like might could, may can, might should, and more.
Aside from the modals, semi-modals, and double modals, the primary auxiliaries are forms of have, be, and do, which are inflected for tense (is versus was, has versus had, do versus did), number (is versus are, has versus have), and person(is versus am versus are, do versus does). These auxiliaries help to indicate verbal nuances like emphasis, the perfect and progressive aspects, and the passive voice. Here are some examples, adapted from Ernest Hemmingway’s The Old Man and the Sea:
Those who did catch sharks had taken them to the shark factory on the other side of the cove … (emphatic do and perfect aspect had)
The old man opened his eyes and for a moment he was coming back from a long way away. (progressive aspect)
His shirt was patched so many times that it was like the sail … (passive voice)
The primary auxiliaries come before the negative adverb not and allow contraction to it.
They didn’t catch sharks.
His shirt wasn’t patched.
He hadn’t taken the sharks.
And they play a role in questions by hopping to the left over the subject
Did they catch sharks?
Was his shirt patched?
Had he taken the sharks?
or by being copied at the end in a tag question.
They caught sharks, didn’t they?
His shirt was patched, wasn’t it?
He had taken the sharks, hadn’t he?
Main verbs like see and go and walk don’t do any of those tricks.
Things get even curiouser, however, because the helping verbs have and do have doppelgangers that actually are main verbs.
The old man did his chores.
His shirt had a tear in it.
How do we know these are main verbs and not helping verbs? Well, for one thing, they are the only verbs in the sentence. For another, they can occur with other helping verbs:
The old man had done his chores.
His shirt had had a tear in it all day.
And if you make the sentences questions or negate them, you have to add a form of auxiliary do.
Did the old man do his chores?
Did his shirt have a tear in it?
The helping verb be also has a doppelganger main verb, but the forms of main verb be behave pretty much just like the helping verb. More curious behavior, keeping us on our toes. The first sentence below has past tense main verb was followed by an adjective; the other two have the past tense helping verb was.
The shark was tenacious. (main verb was)
The shark was never caught. (auxiliary was)
The old man was trying his best. (auxiliary was)
But all three was forms hop to the left in questions.
Was the shark tenacious?
Was the shark ever caught?
Was the old man trying his best?
The curious behavior of helping verbs goes on and on, with different dialects doing different things. If you’ve read many British novels or watched British television you might have noticed forms of helping verb do popping up in elliptical sentences. Here’s an example from J. R. R. Tolkien’s The Two Towers: “Sam frowned. If he could have bored holes in Gollum with his eyes, he would have done.” (For a study of these forms, check out Ronald Butters’s 1983 article “Syntactic change in British English propredicates.”)
In African American English, the auxiliary done lends a completive meaning to events. You can see it in these dialogue examples from August Wilson’s Fences and from Walter Mosely’s Blond Faith: “Now I done give you everything I got to give you!” and “Didn’t she tell you that Pericles done passed on.” For more on this use of done, take a look at the chapters by Lisa J. Green and Walter Sistrunk and by Charles E. DeBose in the Oxford Handbook of African American Language.
We’ve just scratched the surface of auxiliaries. I hope you’ve become curious about these curious words.
Featured image by Alexander Grey via Unsplash (public domain)
“Lying” in computer-generated texts: hallucinations and omissions
There is huge excitement about ChatGPT and other large generative language models that produce fluent and human-like texts in English and other human languages. But these models have one big drawback, which is that their texts can be factually incorrect (hallucination) and also leave out key information (omission).
In our chapter for The Oxford Handbook of Lying, we look at hallucinations, omissions, and other aspects of “lying” in computer-generated texts. We conclude that these problems are probably inevitable.
Omissions are inevitable because a computer system cannot cram all possibly-relevant information into a text that is short enough to be actually read. In the context of summarising medical information for doctors, for example, the computer system has access to a huge amount of patient data, but it does not know (and arguably cannot know) what will be most relevant to doctors.
Hallucinations are inevitable because of flaws in computer systems, regardless of the type of system. Systems which are explicitly programmed will suffer from software bugs (like all software systems). Systems which are trained on data, such as ChatGPT and other systems in the Deep Learning tradition, “hallucinate” even more. This happens for a variety of reasons. Perhaps most obviously, these systems suffer from flawed data (e.g., any system which learns from the Internet will be exposed to a lot of false information about vaccines, conspiracy theories, etc.). And even if a data-oriented system could be trained solely on bona fide texts that contain no falsehoods, its reliance on probabilistic methods will mean that word combinations that are very common on the Internet may also be produced in situations where they result in false information.
Suppose, for example, on the Internet, the word “coughing” is often followed by “… and sneezing.” Then a patient may be described falsely, by a data-oriented system, as “coughing and sneezing” in situations where they cough without sneezing. Problems of this kind are an important focus for researchers working on generative language models. Where this research will lead us is still uncertain; the best one can say is that we can try to reduce the impact of these issues, but we have no idea how to completely eliminate them.
“Large generative language models’ texts can be factually incorrect (hallucination) and leave out key information (omission).”
The above focuses on unintentional-but-unavoidable problems. There are also cases where a computer system arguably should hallucinate or omit information. An obvious example is generating marketing material, where omitting negative information about a product is expected. A more subtle example, which we have seen in our own work, is when information is potentially harmful and it is in users’ best interests to hide or distort it. For example, if a computer system is summarising information about sick babies for friends and family members, it probably should not tell an elderly grandmother with a heart condition that the baby may die, since this could trigger a heart attack.
Now that the factual accuracy of computer-generated text draws so much attention from society as a whole, the research community is starting to realize more clearly than before that we only have a limited understanding of what it means to speak the truth. In particular, we do not know how to measure the extent of (un)truthfulness in a given text.
To see what we mean, suppose two different language models answer a user’s question in two different ways, by generating two different answer texts. To compare these systems’ performance, we would need a “score card” that allowed us to objectively score the two texts as regards their factual correctness, using a variety of rubrics. Such a score card would allow us to record how often each type of error occurs in a given text, and aggregate the result into an overall truthfulness score for that text. Of particular importance would be the weighing of errors: large errors (e.g., a temperature reading that is very far from the actual temperature) should weigh more heavily than small ones, key facts should weigh more heavily than side issues, and errors that are genuinely misleading should weigh more heavily than typos that readers can correct by themselves. Essentially, the score card would work like a fair school teacher who marks pupils’ papers.
We have developed protocols for human evaluators to find factual errors in generated texts, as have other researchers, but we cannot yet create a score card as described above because we cannot assess the impact of individual errors.
What is needed, we believe, is a new strand of linguistically informed research, to tease out all the different parameters of “lying” in a manner that can inform the above-mentioned score cards, and that may one day be implemented into a reliable fact-checking protocol or algorithm. Until that time, those of us who are trying to assess the truthfulness of ChatGPT will be groping in the dark.
Featured image by Google DeepMind Via Unsplash (public domain)
Etymologists interacting with the public on a day-to-day basis usually receive questions about words like copacetic and shenanigans, but so many nouns, verbs, adjectives, adverbs, and even prepositions and conjunctions not crying for attention are not less, perhaps even more, interesting. Over the years, I have written about summer, winter, ice, sheep, dog, live, leave, good, bad, red, and even such inconspicuous words as but and yet, to mention just a few. We use them like old trusty tools and never stop to ask where they came from. Somebody somewhere coined them, which probably means that once upon a time they were as transparent to speakers as giggle and hiccup. But today their origin is either debatable or unknown. Isn’t a dog called a dog because it looks like a dog, runs like a dog, and barks like a dog? This is what the naïve speaker thinks, though nothing in the group d-o-g suggests a muzzle, swiftness, or any kind of explosive sound. How did day and night get their names? Aren’t the words in some way associated with light and darkness? Dictionaries know a lot about the oldest history of both but do not always provide a clue to the “motivation” that might explain their origin.
Let us look at day. Its past is not totally hidden (no reference to or pun on clear obscure!). Day has exact cognates everywhere in Old Germanic, including of course Old English, as well as in Sanskrit, Celtic, Slavic, and elsewhere. In Old Icelandic, the proper name Dagr has been recorded. Similar names existed in Gothic and Old High German. Does this fact testify to the word’s significance in some religious ritual? We’ll never know. What conclusion will an etymologist two thousand years from now draw about our names June, Melody, and Makepeace? Day, it should be remembered, does not always mean “a period of twenty-four hours” (as in a few days ago) or half of this period (as in day and night or daytime): it sometimes refers to “a certain period or date” as in Doomsday, the day of reckoning, I’ll remember it until my dying day, and the like. Even if you were born at night, you probably celebrate your birthday. It follows that we are not quite sure where to begin our exploration, though day as “the period of light” looks more promising: after all, law-abiding citizens tend to make their arrangements for the time when there is enough light around, while at night most of us sleep.
Speakers of Modern English no longer realize that dawn has the same root as day. The verb to dawn means “to begin to grow light,” and the same reference is obvious in the phrase it dawned on me. I’ll skip the phonetic part of the story (why dawn? In German, unlike what we observe in English, the connection between the noun Tag and the verb tagen is immediately obvious). Will then a search for the etymology of day take us to the idea of “light,” as suggested above? Perhaps. But first, let us remember Latin diēs “day.” Its root occurs in the English words diurnal, dial “an instrument to tell the time of day by the shadow cast by the sun,” and diary, all three of course borrowed from Romance. Though diēs and day sound somewhat alike, they are not related (a fact often mentioned in dictionaries, to warn readers against what looks like an obvious conclusion), and as though to prove the absence of ties between them, language provides us with a word like Gothic sin-tiens “daily,” in which sin– means “one” and –tien– is a cognate of the Latin word. The correspondence d (Latin) ~ t (Gothic or any other Germanic language) is regular, by the so-called First Consonant Shift: compare Latin duo versus English two.
This –tien- has unquestionable correspondences all over the Indo-European world: for example, Sanskrit had dínam “day.” In Latin, we find the word nun-dinum “market held every ninth day.” Also, Russian den’ “day” (with cognates elsewhere in Slavic) and many other references elsewhere to burning, ashes, and warmth belongs here too. On the strength of, among others, several Greek words meaning “appear” and “visible,” the root of all such words has been understood as “shining.” Since the Sanskrit dèvas (obviously related) means “god,” this idea looks realistic. The Indo-Europeans habitually referred to “god” as “shining” or “sky” (such was Latin Jū-piter “sky father,” known to the ancient Scandinavians as Týr, no longer a sky god; but the name reveals his distant past). Yet it is still odd that both the words related to day and those related to diēs, though unconnected, sound somewhat alike and not only mean “day” but also begin with d. Did d suggest burning, heat, or glowing or refer to things dry and arid? Such ancient sound-symbolic associations are beyond reconstruction. They are often hard to pinpoint even in our modern languages.
Another puzzling lookalike is Sanskrit áhar “day.” It almost rhymes with Proto-Germanic dagaz but lacks d-, to which, above, I ventured to ascribe magical properties. An incredible coincidence? The Sanskrit noun has no correspondences in Germanic, Romance, Celtic, and elsewhere. (At least, none has been discovered.) Did áhar once have d- and lose it? The fertile imagination of historical linguists reconstructed several processes that could be responsible for the loss. Initial d- does sometimes disappear for no known reasons. For instance, t in tear (from the eye), which, as expected, corresponds to d outside Germanic, is sometimes absent altogether: this is true of Sanskrit and Baltic, among others. We have enough trouble with s–mobile (it tends to turn up wherever it wants). Did d–mobile also exist? Most unlikely.
My point has been to show how intriguing some of our common words sometimes are. Copacetic is late (it was first recorded in the twentieth century), while day is older than most of the hills around us. But the problem of origin remains the same: people coin words and etymologists wander in a labyrinth of look-alikes, roots, fleeing initial, and final consonants, and emerge with the all-too familiar verdict: “Origin unknown (uncertain, disputed).” Yet day probably did refer to heat or a bright light. This conclusion sounds reasonable, assuming (and this a reliable assumption) that the word’s initial sense was “the time of light,” rather than “a certain period, date.” Plain as day? Almost.
Featured image by Ivana Cajina via Unsplash (public domain)
Real patterns and the structure of language
There’s been a lot of hype recently about the emergence of technologies like ChatGPT and the effects they will have on science and society. Linguists have been especially curious about what highly successful large language models (LLMs) mean for their business. Are these models unearthing the hidden structure of language itself or just marking associations for predictive purposes?
In order to answer these sorts of questions we need to delve into the philosophy of what language is. For instance, if Language (with a big “L”) is an emergent human phenomenon arising from our communicative endeavours, i.e. a social entity, then AI is still some ways off approaching it in a meaningful way. If Chomsky, and those who follow his work, are correct that language is a modular mental system innately given to human infants and activated by miniscule amounts of external stimulus, then AI is again unlikely to be linguistic, since most of our most impressive LLMs are sucking up so many resources (both in terms of data and energy) that they are far from this childish learning target. On the third hand, if languages are just very large (possibly infinite) collections of sentences produced by applying discrete rules, then AI could be super-linguistic.
In my new book, I attempt to find a middle ground or intersection between these views. I start with an ontological picture (meaning a picture of what there is “out there”) advocated in the early nineties by the prominent philosopher and cognitive scientist, Daniel Dennett. He draws from information theory to distinguish between noise and patterns. In the noise, nothing is predictable, he says. But more often than not, we can and do find regularities in large data structures. These regularities provide us with the first steps towards pattern recognition. Another way to put this is that if you want to send a message and you need the entire series (string or bitmap) of information to do so, then it’s random. But if there’s some way to compress the information, it’s a pattern! What makes a pattern real, is whether or not it needs an observer for its existence. Dennett uses this view to make a case for “mild realism” about the mind and the position (which he calls the “intentional stance”) we use to identify minds in other humans, non-humans, and even artifacts. Basically, it’s like a theory we use to predict behaviour based on the success of our “minded” vocabulary comprising beliefs, desires, thoughts, etc. For Dennett, prediction matters theoretically!
If it’s not super clear yet, consider a barcode. At first blush, the black lines of varying length set to a background of white might seem random. But the lines (and spaces) can be set at regular intervals to reveal an underlying pattern that can be used to encode information (about the labelled entity/product). Barcodes are unique patterns, i.e. representations of the data from which more information can be drawn (by the way Nature produces these kinds of patterns too in fractal formation).
“The methodological chasm between theoretical and computational linguistics can be surmounted.”
I adapt this idea in two ways in light of recent advances in computational linguistics and AI. The first reinterprets grammars, specifically discrete grammars of theoretical linguistics, as compression algorithms. So, in essence, a language is like a real pattern. Our grammars are collections of rules that compress these patterns. In English, noticing that a sentence is made up of a noun phrase and verb phrase is such a compression. More complex rules capture more complex patterns. Secondly, discrete rules are just a subset of continuous processes. In other words, at one level information theory looks very statistical while generative grammar looks very categorical. But the latter is a special case of the former. I show in the book how some of the foundational theorems of information theory can be translated to discrete grammar representations. So there’s no need to banish the kinds of (stochastic) processes often used and manipulated in computational linguistics, as many theoretical linguists have been wont to do in the past.
This just means that the methodological chasm between theoretical and computational linguistics, which has often served to close the lines of communication between the fields, can be surmounted. Ontologically speaking, languages are not collections of sentences, minimal mental structures, or social entities by themselves. They are informational states taken from complex interactions of all of the above and more (like the environment). On this view, linguistics quickly emerges as a complexity science in which the tools of linguistic grammars, LLMs, and sociolinguistic observations all find a homogeneous home. Recent work on complex systems, especially in biological systems theory, has breathed new life into this interdisciplinary field of inquiry. I argue that the study of language, including the inner workings of both the human mind and ChatGPT, belong within this growing framework.
For decades, computational and theoretical linguists have been talking different languages. The shocking syntactic successes of modern LLMs and ChatGPT have forced them into the same room. Realising that languages are real patterns emerging from biological systems gets someone to break the awkward silence…
Featured image by Google DeepMind Via Unsplash (public domain)
Real patterns and the structure of language
There’s been a lot of hype recently about the emergence of technologies like ChatGPT and the effects they will have on science and society. Linguists have been especially curious about what highly successful large language models (LLMs) mean for their business. Are these models unearthing the hidden structure of language itself or just marking associations for predictive purposes?
In order to answer these sorts of questions we need to delve into the philosophy of what language is. For instance, if Language (with a big “L”) is an emergent human phenomenon arising from our communicative endeavours, i.e. a social entity, then AI is still some ways off approaching it in a meaningful way. If Chomsky, and those who follow his work, are correct that language is a modular mental system innately given to human infants and activated by miniscule amounts of external stimulus, then AI is again unlikely to be linguistic, since most of our most impressive LLMs are sucking up so many resources (both in terms of data and energy) that they are far from this childish learning target. On the third hand, if languages are just very large (possibly infinite) collections of sentences produced by applying discrete rules, then AI could be super-linguistic.
In my new book, I attempt to find a middle ground or intersection between these views. I start with an ontological picture (meaning a picture of what there is “out there”) advocated in the early nineties by the prominent philosopher and cognitive scientist, Daniel Dennett. He draws from information theory to distinguish between noise and patterns. In the noise, nothing is predictable, he says. But more often than not, we can and do find regularities in large data structures. These regularities provide us with the first steps towards pattern recognition. Another way to put this is that if you want to send a message and you need the entire series (string or bitmap) of information to do so, then it’s random. But if there’s some way to compress the information, it’s a pattern! What makes a pattern real, is whether or not it needs an observer for its existence. Dennett uses this view to make a case for “mild realism” about the mind and the position (which he calls the “intentional stance”) we use to identify minds in other humans, non-humans, and even artifacts. Basically, it’s like a theory we use to predict behaviour based on the success of our “minded” vocabulary comprising beliefs, desires, thoughts, etc. For Dennett, prediction matters theoretically!
If it’s not super clear yet, consider a barcode. At first blush, the black lines of varying length set to a background of white might seem random. But the lines (and spaces) can be set at regular intervals to reveal an underlying pattern that can be used to encode information (about the labelled entity/product). Barcodes are unique patterns, i.e. representations of the data from which more information can be drawn (by the way Nature produces these kinds of patterns too in fractal formation).
“The methodological chasm between theoretical and computational linguistics can be surmounted.”
I adapt this idea in two ways in light of recent advances in computational linguistics and AI. The first reinterprets grammars, specifically discrete grammars of theoretical linguistics, as compression algorithms. So, in essence, a language is like a real pattern. Our grammars are collections of rules that compress these patterns. In English, noticing that a sentence is made up of a noun phrase and verb phrase is such a compression. More complex rules capture more complex patterns. Secondly, discrete rules are just a subset of continuous processes. In other words, at one level information theory looks very statistical while generative grammar looks very categorical. But the latter is a special case of the former. I show in the book how some of the foundational theorems of information theory can be translated to discrete grammar representations. So there’s no need to banish the kinds of (stochastic) processes often used and manipulated in computational linguistics, as many theoretical linguists have been wont to do in the past.
This just means that the methodological chasm between theoretical and computational linguistics, which has often served to close the lines of communication between the fields, can be surmounted. Ontologically speaking, languages are not collections of sentences, minimal mental structures, or social entities by themselves. They are informational states taken from complex interactions of all of the above and more (like the environment). On this view, linguistics quickly emerges as a complexity science in which the tools of linguistic grammars, LLMs, and sociolinguistic observations all find a homogeneous home. Recent work on complex systems, especially in biological systems theory, has breathed new life into this interdisciplinary field of inquiry. I argue that the study of language, including the inner workings of both the human mind and ChatGPT, belong within this growing framework.
For decades, computational and theoretical linguists have been talking different languages. The shocking syntactic successes of modern LLMs and ChatGPT have forced them into the same room. Realising that languages are real patterns emerging from biological systems gets someone to break the awkward silence…
Featured image by Google DeepMind Via Unsplash (public domain)
Confronting <em>bud</em> and <em>buddy</em>
In the previous installment (14 June 2023), I mentioned several attempts to explain the origin of bud(dy). See also the comment at the bottom of that post. It may perhaps be useful to remember that monosyllabic words beginning with b, d, g ~ p, t, k and ending in one of those consonants (bed, big, pig, kick, gig, dog, gag, keg, dab, bug, and a host of others) are notoriously obscure from the etymological point of view. Sound imitative? Sound-symbolic? Baby words? Borrowed? Some of them once had an ending (e, i, a), which is of no consequence as regards their origin. Incidentally, bud (on a plant) also poses problems. Known from texts since the Middle English period, it has a Dutch lookalike with t at the end and resembles French bouton, which may be of Germanic origin, but this derivation is far from certain. In one widely-used but rather unreliable dictionary, buddy is said to be “etymologically identical with the adjective buddy ‘full of buds’.” Thus, our correspondent, who had a similar idea, even though she cannot be said to be in good company, is at least not alone. Another complicating factor deserves mention. Such nouns and verbs may be coined, forgotten, and coined again in the same form. After all, it is not too hard to come up with words like bob, gab, pad, and so forth.
Buddy poses familiar problems. It may or may not be a native word. Everybody is agreed that we are dealing with an Americanism. It appeared in texts around the year 1800, which excludes the idea of its derivation from bud “part of a plant.” Strange things sometimes happen. Guy, one the most common American words, goes back to a proper name and the 1605 Gunpowder Plot. The plot happened in England, but guy ousted or limited the use of pal, fellow, and their likes in American English. Last week, I mentioned Skeat’s derivation of buddy from booty-fellow and expressed my doubts on this score. Despite my admiration of everything Walter W. Skeat did, I keep thinking that in this case he was wrong.
What follows depends largely on Jeremy Bergerson’s article in the Dutch periodical Leuvense Bijdragen (91, 2002, 63-71). Dutch has the dialectal word boetje “brother.” It has been recorded in numerous variants, and, according to a reasonable suggestion, it is an affectionate form of the baby word boe “brother.” In recent English scholarship, the idea that buddy goes back to brother has gained the upper hand. The direction seems to be correct, but the way may have been less straight than etymologists would like it. The problem is the loss of r after b. In languages in which r is a trill (or a roll: both terms mean the same), for instance, Spanish, Italian, and Russian (to cite a few examples), r is usually the hardest sound to master. Children tend to substitute l for it. But English r is NOT a trill, and the well-known substitution for it is w. Many of us have heard children say: “I am hungwy.” This is also a much-ridiculed pseudo-aristocratic affectation. As a general rule, an English-speaking baby would probably not say buddy for brother.
We may try to ask for help abroad. Dutch bout, beut, boetje, and budde, among many other similar forms, seem to go back to the baby word boe (pronounced with a vowel like English oo in boo). The story is partly reminiscent of the history of English boy. Boy, too, may be a derivative of a baby word for “little brother,” and the Old English name Boia perhaps contains the “root” in its pristine form. Dutch Boio, Boiga, Boga, and even Scandinavian Bo may once have meant “little brother.” (More details on this score can be found in my and J. Lawrence Mitchell’s An Analytic Dictionary of English Etymology: An Introduction, pp. 15-16). If this reconstruction is correct, the many words cited above do go back to “brother,” but not to the word brother. The baby etymon contained no r. This suggestion runs counter to what we find in some of our most respectable sources.
Unexpected light on the origin of buddy comes from the English word Boots. The word is known from such compounds as slyboots and lazy–boots, but most will probably remember Boots “a name for the servant in hotels who cleans the boots.” The OED cites numerous examples. Apparently, Boots emerged with the meaning “servant; a person at the lowest level of the hierarchy” rather than “shoeshine boy.” This is evidenced by such senses as “the youngest officer in a regiment” and others. The hotel Boots did clean boots and put them outside the gentleman’s door, but this is not why he received his name. George Webbe Dasent used the word Boots for rendering Norwegian Askepott ~ Askeladden (this is what the despised third son in fairytales, “male Cinderella,” is called). If Dasent had associated Askepott only or mainly with the shoe shiner, he would hardly have used the word in such a context. An ingenious correspondent to Notes and Queries once suggested that Puss in Boots might be a misnomer, because allegedly, Puss and Boots, that is, Puss as Servant was meant. But the English title is a translation from French, where Chat botté is unambiguous.
Bergerson suggested that Boots is a borrowing of Dutch boet, a childish word for “brother.” We find Dutch boet “a recent recruit in the Indian Army” and a few other words that fit our story well. If boet (pronounced approximately like English boot) was indeed borrowed into English with the meaning “a person of inferior rank,” the rest is plain sailing. The ending s is added freely to English words. I, for example, have dealt with two guineapigs: Cuddles and Sniffers. In Dickens’s Dombey and Son, the servant who wheeled Mrs Skewton around had the name Withers. Dobbin of ours, a most important character in Thackeray’s Vanity Fair, was called Figs, because his father sold all kinds of groceries. And don’t we occasionally live in digs and have things for keeps?
Dutch words in English are numerous (very numerous indeed). If Boots is one of them, perhaps so is buddy from boetje. The voicing of t between vowels is no problem. Buddy is an American coinage, and in American English, intervocalic t is voiced, so that seated and seeded, Plato and playdough, sweetish and Swedish, futile and feudal (to mention just a few examples), are homophones pairwise. With time, the vowel in boetje acquired, as expected, the value of u in but, and butty ~ buddy was born. Speakers quite correctly interpreted the last sound as a diminutive suffix and produced bud from buddy by back formation. The word has just the expressive value one needs: compare studs “a great virile guy,” as in Studs Lonigan byJames T. Farrell.
Have we solved the riddle? No, the etymology of buddy and Boots will probably remain “debatable” (unless it suddenly gains universal recognition—a rare case in etymological studies), but perhaps we have made a step in the right direction.
Featured image via pxfuel (public domain)
The company we keep, part two: <em>bud(dy)</em>
I am picking up where I left off two weeks ago. Since there have been no comments or letters connected with part one, I will, as promised, tackle the convoluted history of bud. The Internet is awash in suggestions about the origin of this universally known word. Some such suggestions are reasonable, even clever; others fanciful (for instance, bud has been derived by some from Pashto, from Spanish, and even from the names Buddha and Budweiser!). Of course, we now have the opinion of the OED online, and yet I may add something to the ongoing discussion—no revolutionary hypothesis: rather a glance at the state of the art. Predictably, bud ~ buddy can also be found in the first volume of the original OED, but its editor James A. H. Murray was reticent on the word’s etymology. Today’s editors have much more to say about the history of bud; yet the sought-for origin remains disputable and will probably remain such.
Though the first volume of the OED appeared in 1896, people kept returning to the history of bud, and some of their hypotheses deserve attention. Perhaps the most important part of the tale is whether bud ~ buddy has anything to do with the word butty “comrade.” Here are some quotations from the letters sent to the biweekly Notes and Queries (my constant and inexhaustible source of inspiration). “What is the origin of the word butty, gamekeeper’s slang for “comrade”? The dog was took away home to granny by my butty” (Richard Jefferies, ‘The Amateur Poacher,’ 1896, p. 117”). The editor answered:
“The origin is unknown. The ‘H.E.D.’ [that is, Historical English Dictionary, the original name of the OED; even Murray could not remember when HED became OED!] says that it is a possible corruption of booty. The word is in general use in England. See ‘English Dialect Dictionary’ [by Joseph Wright].”
Other contributors also cited butty from the regions where they lived, and Walter W. Skeat, ever-ready with a suggestion when it came to etymology, wrote:
I would suggest that butty, comrade, is a mere abbreviation of booty-fellow, one who shares in booty; hence a comrade. The full form occurs in Palsgrave [an early sixteenth-century lexicographer] and is duly explained in the “H.E.D.”, s. v. [sub voce “under the word”] “Booty,” §5.
Since that time the derivation of butty from booty-fellow has been repeated more than once. Other contributors to Notes and Queries mentioned butty collier “a man who contracts with the proprietor of a coalpit to get and raise coal to the bank at so much per ton.” In that context, butty was also connected with boot. Additionally, “in ironworks, where two men frequently manage a forge, one superintendent by day and the other by night, each often describes the other as his butty.” In the same region, “a man and a woman living together irregularly sometimes describe each other as his or her butty, and other people would so describe their relations” (1901). “In Warwickshire [the West Midlands] sweethearts who keep company describe their association as buttying with each other.” Bud “husband” has been used for centuries, but of course no phonetic legerdemain can produce this wordfrom husband. Finally, in Scotland, buddy, buddie, and butty have been long since used as a pet designation for a little child.
Judging by the meaning of the words mentioned above, it seems that buddy and butty have rather little to do with booty, a French word of Germanic origin. Booty-fellow is close but seems to have little to do with buddy as we know it. Below, I will quote part of a letter from the journal American Speech 4, 1929, p. 389. Though the author’s suggestion looks like an example of folk etymology, bud is such a controversial word that even dubious remarks have some value.
The Standard Dictionary [that is, Funk and Wagnalls] gives butty as a variant of buddy. I raise the query whether precedence should not be the other way. When I was a boy in the Pennsylvania coalfields, one of the commonest words in the miners’ speech [again mining, as in England!] was butty, or buttie, as I have always preferred to spell it, meaning work-fellow. In the cramped, underground and honey-combs where bituminous coal is dug, the man with a pick and the man with the shovel are literally butties, working all day long buttock to buttock, or in the vulgar but comradely abbreviation, butt to butt. The word was used generally in a technical sense, without any endearing connotation of spiritual nearness. But the boys of the community thirty years ago adopted the word for the idea which later at school they were to express by chum, a word which I confess always sounds emasculated and utterly lacking the vulgar warmth of buttie (!).
The author begins his letter by saying that buddy “is by origin a childish corruption of brother, with a familiar form ‘bud.’ … it is incontrovertible that bud and buddy are diminutives of brother…”.
The “emasculated” chum was discussed briefly and inconclusively in the previous blog post, and I will not comment on the author’s use of the word incontrovertible. I have more than once expressed my objections to terms like certainly, undoubtedly, obviously, and so forth. (Very few things in etymology are incontrovertible.) I am only a bit confused about the order of events. If buddy is from brother, how can it a variant of butty? Here, I’ll also mention the enigmatic sixteenth-century adjective baddy of unclear meaning, but, apparently, with some negative connotations. This word deserves attention, because Murray devoted a long letter to “the queer phrase” paddy persons in Notes and Queries 9/XII, 1903, 87-88, and nothing that great man wrote should fall through the cracks. It turned out that paddy persons is a misreading: the correct 1585 phrase is baddy persons, and I wondered whether baddy has anything to do with buddy. My question remains unanswered, but the syntax of the quotation is characteristic, and I would like to quote that phrase as a postscript to my blog post.
Here is this phrase: “I doubt [= fear] not that the flower of the pressed English bandes are gone, and that the remnant supplied with such baddy persons as commonly, in voluntary procurements, men are glad to accept.” Baddy will, unfortunately, remain a minor riddle, but another thing should not be overlooked: “The flower of such band(es) are gone.” The subject is of course flower, and the verb should have been is. Yet the writer made the verb agree with the noun closest to it (bands). As is well-known, colonial languages are conservative. American English is no exception. The Revised Version of the Bible has: “Our Father which art in heaven.” And I sometimes hear statements like: “That’s the guy which I told you about.” But this usage is relatively rare. By contrast, the syntax of the phrase quoted above is ineradicable: students’ papers teem with it. I have once quoted my favorite example from an undergraduate’s essay: “The mood of the tales are gloomy.” This usage, well-known to the historians of English syntax, is still rather common in Ireland and in other “colonial” varieties of English. It would be interesting to heart what our readers know about it.
Next week, I’ll finish my discussion of buddy and say what little I think I know about its origin. Nothing in my suggestion will be original or incontrovertible.
Featured image via Wikimedia Commons (public domain)
What can Large Language Models offer to linguists?
It is fair to say that the field of linguistics is hardly ever in the news. That is not the case for language itself and all things to do with language—from word of the year announcements to countless discussions about grammar peeves, correct spelling, or writing style. This has changed somewhat recently with the proliferation of Large Language Models (LLMs), and in particular since the release of OpenAI’s ChatGPT, the best-known language model. But does the recent, impressive performance of LLMs have any repercussions for the way in which linguists carry out their work? And what is a Language Model anyway?
At heart, all an LLM does is predict the next word given a string of words as a context —that is, it predicts the next, most likely word. This is of course not what a user experiences when dealing with language models such as ChatGPT. This is on account of the fact that ChatGPT is more properly described as a “dialogue management system”, an AI “assistant” or chatbot that translates a user’s questions (or “prompts”) into inputs that the underlying LLM can understand (the latest version of OpenAI’s LLM is a fine-tuned version of GPT-4).
“At heart, all an LLM does is predict the next word given a string of words as a context.”
An LLM, after all, is nothing more than a mathematical model in terms of a neural network with input layers, output layers, and many deep layers in between, plus a set of trained “parameters.” As the computer scientist Murray Shanahan has put it in a recent paper, when one asks a chatbot such as ChatGPT who was the first person to walk on the moon, what the LLM is fed is something along the lines of:
Given the statistical distribution of words in the vast public corpus of (English) text, what word is most likely to follow the sequence “The first person to walk on the Moon was”?
That is, given an input such as the first person to walk on the Moon was, the LLM returns the most likely word to follow this string. How have LLMs learned to do this? As mentioned, LLMs calculate the probability of the next word given a string of words, and it does so by representing these words as vectors of values from which to calculate the probability of each word, and where sentences can also be represented as vectors of values. Since 2017, most LLMs have been using “transformers,” which allow the models to carry out matrix calculations over these vectors, and the more transformers are employed, the more accurate the predictions are—GPT-3 has some 96 layers of such transformers.
The illusion that one is having a conversation with a rational agent, for it is an illusion, after all, is the result of embedding an LLM in a larger computer system that includes background “prefixes” to coax the system into producing behaviour that feels like a conversation (the prefixes include templates of what a conversation looks like). But what the LLM itself does is generate sequences of words that are statistically likely to follow from a specific prompt.
It is through the use of prompt prefixes that LLMs can be coaxed into “performing” various tasks beyond dialoguing, such as reasoning or, according to some linguists and cognitive scientists, learn the hierarchical structures of a language (this literature is ever increasing). But the model itself remains a sequence predictor, as it does not manipulate the typical structured representations of a language directly, and it has no understanding of what a word or a sentence means—and meaning is a crucial property of language.
An LLM seems to produce sentences and text like a human does—it seems to have mastered the rules of the grammar of English—but at the same time it produces sentences based on probabilities rather on the meanings and thoughts to express, which is how a human person produces language. So, what is language so that an LLM could learn it?
“An LLM seems to produce sentences like a human does but it produces them based on probabilities rather than on meaning.”
A typical characterisation of language is as a system of communication (or, for some linguists, as a system for having thoughts), and such a system would include a vocabulary (the words of a language) and a grammar. By a “grammar,” most linguists have in mind various components, at the very least syntax, semantics, and phonetics/phonology. In fact, a classic way to describe a language in linguistics is as a system that connects sound (or in terms of other ways to produce language, such as hand gestures or signs) and meaning, the connection between sound and meaning mediated by syntax. As such, every sentence of a language is the result of all these components—phonology, semantics, and syntax—aligning with each other appropriately, and I do not know of any linguistic theory for which this is not true, regardless of differences in focus or else.
What this means for the question of what LLMs can offer linguistics, and linguists, revolves around the issue of what exactly LLMs have learned to begin with. They haven’t, as a matter of fact, learned a natural language at all, for they know nothing about phonology or meaning; what they have learned is the statistical distribution of the words of the large texts they have been fed during training, and this is a rather different matter.
As has been the case in the past with other approaches in computational linguistics and natural language processing, LLMs will certainly flourish within these subdisciplines of linguistics, but the daily work of a regular linguist is not going to change much any time soon. Some linguists do study the properties of texts, but this is not the most common undertaking in linguistics. Having said that, how about the opposite question: does a run-of-the-mill linguist have much to offer to LLMs and chatbots at all?
Featured image: Google Deepmind via Unsplash (public domain)
During the news coverage of the COVID pandemic, I enjoyed seeing Dr Anthony Fauci on television and hearing his old-school Brooklyn accent, still shining through in his late seventies in words like Pfizer, because, data, here, and that.
But my favorite expression to listen for was his use of “down the pike” to mean “in the future.” Fauci explained once that “you don’t see for days or weeks down the pike.” Another time he said “before you know it, two to three weeks down the pike, you’re in trouble.” Discussing vaccine testing he said “So we go into phase one, it’ll take about three months to determine if it’s safe. That’ll bring us three or four months down the pike.”
“Down the pike” is an expression I grew up hearing all the time in my home state of New Jersey. And, of course, many people say it besides Anthony Fauci. Joe Biden talks about how government and the private sector should “anticipate and respond to shortages that may be coming down the pike.” If you Google “down the pike” you’ll find it everywhere, even in New York Times headlines like “Is a Trans-Atlantic Pact Coming Down the Pike?” and (with an attempted pun) “Hydrogen Cars, Coming Down the Pike.” You can find occasional instances of things coming “up the pike,” and there is an early twentieth-century slang expression “hit the pike,” meaning “hit the road” or “leave.”
Not everyone is familiar with “down the pike.” Some mislearn it as the semantically plausible “down the pipe” rather than “down the pike.” It’s “down the pike,” but where does it come from?
I had assumed that “down the pike” had something to do with the New Jersey Turnpike, the 117-mile toll highway that runs from New York City to Delaware. The New Jersey Turnpike Authority was created in 1948 and the Turnpike itself was completed in 1951. It’s been collecting tolls ever since, along with the Garden State Parkway, which was completed in 1957.
There are lots of turnpikes, however, and the word goes way back. The Oxford English Dictionary gives examples from the early 1400s. It originally meant “a spiked barrier fixed in or across a road or passage,” and was used as a defense against attacks by men on horseback.
Later turnpike became extended to the sense of a turnstile to block horses. Samuel Johnson offered this definition: “Turnpike… a cross of two bars armed with pikes at the end, and turning on a pin, fixed to hinder horses from entering.” By the late 1600s, the turnpike was a toll booth of sorts, and a late seventeenth-century Act of Parliament refers to “collecting the said [toll]… by setting up a Turnpike or otherwise.”
Turnpike, or the clipping pike, was often used to refer generally to roads in the nineteenth century, and the expression “coming down the pike” was another way of saying “coming down the road.”
By the late 1800s, figurative senses were emerging and taking hold. An 1898 story in the Dayton Herald had the line “Bowling, if dead, is the liveliest corpse that ever ’came down the pike’, as they say on the bowery.” The quote marks suggest that figurative “down the pike” was a newish expression at the time.
Pike was also used as a synonym for boardwalk or midway, and in 1903 the organizers of the St. Louis World’s Fair announced that the upcoming fair would call its promenade “The Pike,” to distinguish it from Chicago’s Midway. Perhaps St. Louis got the idea from Long Beach, California, which debuted a boardwalk amusement zone called “The Pike” even earlier—in 1902. In any event, St. Louis encouraged visitors to “Come Down the Pike,” and there was later a Broadway musical titled “Down the Pike” whose second act took place at the 1904 World’s Fair.
The figurative sense of “coming down the pike” took hold but remained sporadic in the early twentieth century. A 1905 Portland, Maine, newspaper talked about “nothing but anarchy coming down the pike” and “chaos coming down the pike.” A 1936 issue of the International Stereotypers’ and Electrotypers’ Union Journal which refers to “The fall election… coming down the pike.” Both examples are clearly oriented toward time and events rather than space.
By the 1950s the sense of “coming down the pike” to mean happening in the future was increasingly common and it took off in the 1960s and 70s, after the completion of the Interstate Highway System.
With language, you never know what’s coming down the pike.
Featured image: the New Jersey Turnpike, 1992, via Wikimedia Commons (CC BY-SA 2.5)
Observing how various words for “friend” originate and develop is a rather curious enterprise. Some etymologies are trivial, that is, they have been known for a long time and are undisputed. Such is, among others, the case of friend. Any good dictionary will tell the same story. The last two letters of friend (-nd) are a trace of an old present participle. In English, present participles end in –ing (a barking dog, flowering wilderness, and so forth), but at one time, the ending of this part of speech was –and. The other Germanic languages still have some easily recognizable traces of –and (cf. German kommend “coming” and so forth). The root of friend is also transparent to the language historian: it once meant “to love” (so already in Gothic, a fourth-century Germanic language, which has come down to us in a rather full form). In Old English, the word did mean “lover.”
Do we love our relatives? It depends. In a society in which kinship determines people’s behavior, love (the sense of belonging, loyalty, devotion, and many other feelings) has numerous “pre-modern” shades. In any case, Icelandic frændi means neither “friend” nor “lover” but “kinsman, relative.” Those who believe that English spelling will gain greatly if reformed may note that friend and fiend are spelled alike and yet do not rhyme. Fiend is also an old present participle. It once meant “hating; hater.” So much for the origin of friends and friendship.
Another trivial case is fellow. Fellow certainly belongs with the subject announced in the title: compare fellow traveler and good fellow “an affable person.” The word came to English from Scandinavian. Icelandic félagi is a compound: fé means “property, money” (English fee is its distant cousin), while –lagi is related to English lay (as in to lay something). Consequently, a fellow is a person who lays down money with someone in a joint undertaking. The word bears some similarity to companion, except that companion is a borrowing from French, rather than Scandinavian. The Romance roots of this word are obvious: com– “together” and panis “bread” (a companion is therefore “one who breaks bread with another”). My students have never heard the phrase boon companion “good companion.” In older books, this phrase usually means “someone with whom one drinks and makes merry” (for example, we can remember Tony Lumpkin and his boon companions in She Stoops to Conquer), and there is no reason why it should be forgotten.
Sharing food will now take us to the noun mate. We have received and welcomed linguistic guests from Scandinavian and French. It would therefore be wrong to ignore Low German. Enter mate. The word (ge)mate, from gamato-, had the prefix ga- ~ ge-, denoting association, and the root discernible in English meat. Meat once meant “food,” as it still does in sweetmeats and green meat (two other items of the English vocabulary my students fail to recognize), from the root “to cut,” as in German messen and (!) English mete out. Thus, we again end up in a friendly company of food sharers.
A few more notes about this mess. The word mess, of French origin, first meant “a serving of food; dish.” In the eighteenth century, it could refer to “mixed food for an animal.” The familiar sense “medley; jumble; a state of utter disarray” appeared later. The root of mess can be seen in Latin mittere “to send out” (as in e-mit and mess-age). In a way, the noun messmate is a tautological compound like pathway. Both parts mean approximately the same: here “food-food,” rather than “someone with whom ones shares food.”
In the company we keep, three words are more serious etymological puzzles: crony, chum, and buddy.
Crony turned up in books in the 1660s. Samuel Pepys knew the word, and so did Stephen Skinner, the author of the 1671 etymological dictionary of English (the second ever). Both Pepys and Skinner were Cambridge men, and crony, long before it acquired the modem negative connotations, meant “roommate” (at Cambridge). In the entry for 30 May 1665, Pepys speaks of the death of Jack Cole, “who was a great chrony (sic) of mine.” The only reasonable derivation of crony seems to be from Greek Chronios “contemporary,” from Chronos “time.” The use of Greek and Latin words at British universities and public schools was commonplace. Pepys’s spelling of the word with ch probably shows that this is how he too understood the word, even though at that time and later, “elegant” (Greek or Latin) spelling variants were customary. To be sure, by Pepys’s time, the word may have been used for decades and been garbled more than once. Therefore, the current etymology looks acceptable but not certain. The once suggested derivation of crony from some cognate of the verb croon is nonsense.
Chum, the Oxford counterpart of crony, became known from printed texts at almost the same time as crony. (Does it follow that both were, after all, recent seventeenth-century coinages? If so, why did they suddenly come into being?). Both were slang, and those who discussed them two hundred years later sometimes apologized for dealing with such low terms: at that time, slang was a synonym for filth. “I confess I rather like the word, though not a few of those born in the [eighteen]-forties, at least, seem disposed to call it slang” (from an 1895 letter). E. B. Brewer, the author of a once tremendously popular 1870 book on the origin of words and idioms, defined chum as “bedfellow.” It is unclear where he found this gloss, but every two students at old Cambridge and Oxford did indeed share a bed.
Cicero and Tacitus used the Latin word con-taberna-lis (literally, “someone sharing a taberna,” roughly “tavern”) with the sense “comrade” (incidentally, a comrade is also a person sharing a camera “chamber” with someone). But contabernalis could hardly have been “Englished” into chum. Even less probable is the derivation of chum from Latin cum “with.” Those old suggestions deserve little consideration.
In 1896, the great Walter W. Skeat offered his etymology of chum. He cast doubt on the derivation of chum from chamber-(fellow), suggested cautiously in the OED, and indeed the path from chamber to chum “does not run smooth.” Skeat found the following entry in the 1767 very well-known Bremen Dictionary: “[In my translation]: “Kumpan, abbreviated as Kump, associate, companion, comrade; College socius, consors. Engl. chum.” He believed that chum was not only a gloss on but also a borrowing of German Kump and explained how the change may have happened. He reconstructed a devious way from Kump to chum. Yet if people could alter chamber into chum, they could do the same, and much more easily, with Kump. But why should British students have borrowed a German slang term? Though that is again not improbable, the reconstruction remains guesswork. With time, Skeat probably lost enthusiasm for his hypothesis, because he did not even include chum in the latest edition of his Concise Dictionary. Thus, today, we can remain chummy, without knowing why we use the word everybody knows (not a rare case in etymology in its relation to life).
The story of buddy is long, and I’ll leave it for the next blog post, but I hasten to thank our readers for their corrections, conjectures, and comments on cowardly custard. (Sorry for the alliteration!). I also have to inform them that I am off to a conference of the Dictionary Society of North America in Boulder, Colorado. This is my first trip “abroad” since the beginning of the pandemic. Therefore, the buddy thriller will appear a week later than scheduled. The long wait will, I hope, whet everybody’s appetite for more tales in the series of One Thousand and One Etymologies.
Featured image: “Chronos and his child” by Giovanni Francesco Romanelli, via Wikimedia Commons (public domain)
More short words, or negation of the negation
In late spring and summer, when the days are long, I am often overwhelmed by the desire to write something about the shortest words in Modern English. At one time, I discussed if, of, and, in, both, and (very briefly) yet. See the posts for 2 August and 9 August 2017, and 27 July 2022. The reason I am seized by such an unnatural desire is the almost total obscurity of such words. Open the most detailed etymological dictionary of any Indo-European language and look up in, on, at, to, of, and their cognates: you will find long lists of short forms but no origin. Some pronouns, especially she, have been discussed many times, but the discussion invariably turns round their later history.
Now take any of our negations. All over the Indo-European map, the main word begins with n. What is in this sound that invites denial, refutation, or repulsion? Unable to answer this question, I decided to turn to some negative prefixes and began with the word ignoble. Latin nōbilis meant “(well-)known, excellent, highborn” (the range of meanings was broader than in the Modern English adjective noble), while īgnōbilis referred to people of low birth, undistinguished, and insignificant. Note that the first vowel of the Latin word is long (ī). Once, the prefix was the familiar in: compare such adjectives as inadequate, inappropriate, inefficient, and so forth. The original form must have been in-gnōbilis (theroot in it is the same as in gnostic), but in the heavy group ingn-, the first n was lost, and the preceding vowel, by way of compensation (as it were) lengthened. Or perhaps the heavy cluster ngn with its two n’s flanking g was dissimilated. The same sound change accounts of the form of ignorant and ignominious.
In other positions, the process of assimilation is so obvious that we need no reference to historical phonetics to recognize the origins. Such transparent forms are illegal, improper, and irreversible, as opposed to indifferent. Many Latin words ended up in Middle English by way of Old French. That is why enemy (from Old French enemi, with one n) begins with en-, while inimical “hostile” looks perfectly Latin. (An enemy is not an amicus “friend.”)
The sound of n prevails in negations. English once had ne-, corresponding to Latin nē, as in nefarious (from ne– + the root of fas “rejection of divine command”) and in Nemo, familiar thanks to the famous captain: it goes back to nē + homo “nobody.” (Those who remember Jules Verne’s book The Mysterious Island know that the real name of this fictious character was Prince Dakkar.) Being so short, ne tended to attract all kinds of barnacles. Thus, in the earliest Romance languages, p and c clung to it. We can still see the slightly transformed remnant of that c in negate, neglect, and negotiate. The syllable neg- goes back to nec-, and the roots of the aforementioned verbs can be seen in Latin aio “I say,” lego “I gather,” and otium “leisure, ease.” The consonant g is an insertion.
At present, the English negation is no, from nā, an old contraction of ne + ā As though English had too few negations, it borrowed nei “nay” from Scandinavian. There, it was a compound made up of ne + ei “ay, ever,” that is, a reinforced negation. German nein, though it means “no,” corresponds exactly to English no one or none. The word not is a later form of naught, which is the sum of ne and aught (as in for aught I know). Likewise, Russian net “no” is the sum of ne and a short phrase meaning “is here.” The more synonyms we have, the more nuanced our speech becomes and the harder it becomes to make small distinctions. Compare you are not a man and you are no man.
A curious word is German kein, used attributively, for example, kein Mann “no man.” It emerged as nih-ein, occurred also in the form (ni) dehein (with the obscure syllable de-), and meant “some,” rather than “no.” Today, it means only “no,” and if we want to say “no book,” the only way to say it in German is kein Buch. Its k-, from h-, has not been explained to everybody’s satisfaction. Very short words have to fight for their existence and sometimes change according to puzzling rules.
We will probably never know why the sound of n allied itself to negations in Indo-European, and this is a pity, for there must have been something in this consonant that suggested the opposite of unity, attraction, and agreement. Reference to sound symbolism fails us here, as it always does when linguists try to see a direct connection between the articulation of consonants and the meaning of words in which they occur.
However, we observe that n is not the only actor in the game of negation. In Old Icelandic, the adverb ei meant “ever.” English aye (“yes; ever”; cf. the ayes have it) is a borrowing of it. But the other ay meant “not” (thus, simply a verbal sign of reinforcement?) and as a rule, occurred with the particle –gi. Eigi became ekki and, like eigi, stayed in the language. The universally used Scandinavian negation (ei, ej) corresponds exactly to Finnish and Estonian ei, and as could be expected, two hypotheses compete. One defends the Finno-Ugric origin of the Finnish and Estonian word; the other equates it with the Germanic one. In any case, the coincidence is remarkable.
Suffix-like words are called enclitics. Germanic, so rich in negative words, also had several negative enclitics. One of them was –gi, discussed briefly above. It varied with –ki, but the meanings often fluctuated between negative and indefinite, which makes sense, because the line between negation and doubt and even between strict and qualified negation is easy to cross. (Hence the problem of ne and ni in Russian. It drives learners crazy, because those two unstressed words sound very much alike. An even worse horror is that ne and ni are sometimes spelled together with the word they modify, and sometimes separately.)
In Germanic, negative enclitics were rather numerous. Sometimes they corresponded to Latin –que, which meant “and”(!). They abounded in the oldest stages of the language, and poets enjoyed them. One such short particle was Old Icelandic at. Its etymology, cited in dictionaries, carries little conviction, and here we may ignore it and only note that the poets of that period loved multiplying negations. For example, né (é designated a long vowel) might occur before the verb and -at at the end of it. Incidentally, –at also has analogs in Finno-Ugric. The rule prevalent in Modern Germanic that requires only one negation in a sentence is relatively late, and those people who still say I don’t know nothin’ are unaware of the norm imposed on the Standard. Incidentally, in Russian, the English sentence I don’t know anyone anywhere, if translated literally, would sound as “I don’t know nobody nowhere.” Obviously, the more, the better.
Our tour of negations in this blog post was a mere introduction to a complicated subject. The great question is how speakers learned to negate their statements, and we wonder at the variety of the means they invented. We also wonder why the sound n- figures so prominently in the process. Apparently our statement “no means no” had many nuances at the dawn of civilization.
Featured image by cottonbro studio, via Pexels (public domain)
On cowards and custard from a strictly linguistic point of view
It is curious what a multitude of synonyms for “brave” Modern English has (bold, courageous, valiant, fearless, and at least a dozen more), while coward ~ cowardly have practically none. For coward only a few rather unexciting nouns like poltroon and dastard turn up, and for the related adjective we chiefly find compounds like chicken-hearted and faint-hearted. It appears that language endows courage with many shades, while cowardice has only one hue. Such picturesque metaphors as scaredy-cat and milksop don’t count.
The origin of coward is known, and I’ll say nothing about it that cannot be found elsewhere, but the story will take us to other, more interesting words. The strangest thing is that coward, a single common name in English for a person lacking fortitude, is a borrowing from Old French. Where, we would like to know, is the Germanic coward lurking? We’ll discover that ignoble person later but will first note that people seem to have resented this lack of native, grassroots diffidence and mobilized folk etymology to fill the gap. Time and again, coward has been explained as an alteration of cowherd. Why should cowherds be or have been prototypical cowards? The silliness of this derivation did not bother its proponents. Folk etymology seeks an easy explanation rather than logic. The same approach to word origins connected coward with the verb cower, but neither cower nor incidentally, the verb cow has anything to do with coward, even if cower may have reinforced the meaning of our noun.
The source of coward is Old French couard, ultimately, from Latin coda “tail.” In the immensely popular and beautiful poem Reynard, The Fox, translated into Middle English from Old Flemish by the great William Caxton (1415-1490), the hare is called Cuwaert and Kywart. Perhaps the original allusion was to a frightened dog with its tail between its legs. It has also been noted that in heraldry, a lion with its tail between its legs, called couard in Old French, has the deprecatory suffix –ard and means “having a short, drooping, or otherwise ridiculous tail.” The reference to the tail is sometimes called obscure, but on the whole, the existing explanation is not bad. As for the suffix, which we can see even in the name Reynard, it also occurs (among others) in drunkard, dotard, sluggard, and the already mentioned dastard.
Now back to the Germanic coward. We find him in German and Scandinavian: German feige and Dutch veeg mean “cowardly,” while Old Icelandic feigr means “doomed, fated, destined to die.” In Modern Norwegian, Danish, and Swedish, the cognates of feigr mean “fearful; unusual; crazy,” but this change happened under the influence of German. The sense in Old Icelandic is the original one. In German, the transition from “doomed” to “cowardly” went through several stages, one of which was “hated.” The distant origin of our word is unclear and need not bother us. I should only note that the most usual Indo-European root cited in this connection looks unconvincing. The development must have been from “doomed to die” to “afraid of death.” Perhaps one of the intermediate stages was “ready to die.” In some southern German dialects, the related adjective means “almost ripe” and “rotten.” The most ancient root may have had that meaning, as suggested long ago but dismissed by later scholarship without discussion. And this brings us to English fey “fated to die,” an exact cognate of German feig(e), Dutch veeg, and Old Icelandic feigr, now and for many centuries, chiefly Scottish. Such is one of the verbal sources of Germanic fear.
Yet it is not the only one. German Angst “fear” has made its way into English dictionaries, along with the untranslatable Zeitgeist and Schadenfreude. The word, whose specific meaning goes back to Freud, is at least a century old in English, but it still exists only as a synonym for “anxiety.” Itis related to German eng “narrow” and a host of words outside Germanic. When you find yourself in a narrow spot, you experience anxiety, you are afraid. This etymology highlights the development of abstract concepts from concrete ones, about which more will be said below. Here we will only note that the root of Angst can also be seen in anger, anguish, angina, anxious, agnail (!), and so forth. The related words in Greek and Latin sound almost like those in Germanic, and we recognize their English offspring without any trouble.
Fear is also a Germanic word. Its Old English source meant “sudden calamity; danger.” Its cognate in Old Saxon (another West Germanic language) turned up with the sense “ambush.” By contrast, “danger” is the only meaning of Modern German Gefahr and Dutch gevaar. Our knowledge of the subtler meanings of words in the oldest texts is imperfect, because we depend on a limited number of contexts and on occasional, often untrustworthy Latin glosses. Even though such a rigorous semantic law does not exist, it makes sense to assume that abstract senses tend to be derived from more concrete ones. In our case, “ambush” (a source of fear) may have developed into “danger” and “fear.” Yet our material is too scanty for bold generalizations. Though great perils awaited a traveler (ambushes, highwaymen, inhospitable terrain), English fare and its numerous congeners, including German fahren, are not related to fear, because the noun fear had a long vowel in the root, while the vowel in fare and the rest was short. Such niceties did not bother our oldest scholars, but all modern etymological research depends on them.
Here then is a short summary of what we have found. Cowards, those mean-spirited, pusillanimous people who treasured their lives more than the common good, have always existed, but the oldest Germanic words for the lack of bravery seem to derive from the ideas of being fated (doomed, destined) to die. When French words for the spiritual sphere flooded Middle English, the native nouns designating the lack of courage were lost, and coward, a loan from Old French, replaced them. However, fear, an old word, survived, though its original meaning was more concrete, namely, “calamity” or perhaps “accident.” German Angst preserves its status of a foreignism in English, but fear is native and thus makes amends for the imported word coward.
For dessert, I would like to quote the phrase cowardly, cowardly, custard, an alliterative taunt used by boys and reported in 1872 in the biweekly Notes and Queries by a correspondent from Philadelphia. He wrote: “It is supposed to have its origin in the shaking, quivering motion of the confection called ‘custard’. In Microcosmos (1687) [by Johann Martini], Act iii, Tasting says “I have a sort of cowardly custards, born in the city, but bred up at court, that quake foe fear.” The date of the OED’s earliest citation of the phrase is 1833. The explanation given in 1872 seems to be correct: a quivering heart has always been associated with cowardice. In an Old Icelandic poem from the Elder Edda, one of the heroes sees a heart, allegedly cut out of the breast of his brother, but gives the messenger the lie: the heart is quivering. And indeed, the heart belongs to a cook. Then they kill the brother, whose heart looks right! My questions is obvious: Does any of our readers know the taunt? If so, where is it still used?
Featured image: “Two Jack Russell terriers chasing a rabbit into a burrow”, Wellcome Images via Wikimedia Commons (CC BY 4.0)
Etymology gleanings for April 2023
Since my previous attempt to glean on a half-frozen field, I have received quite a few letters and suggestions. All of them deserve a response. But first, as always, let me thank our correspondents for consulting the blog, asking questions, and offering words of encouragement.
Since my subject was hooker “prostitute,” I stayed away from the other senses of the word. Therefore, I apologize to the correspondent who excels in rugby, calls himself a proud hooker, and found nothing about the game in my post. I share his pride, feel his pain, and have two things to say. First, a hooker is obviously anyone who wields a hook. In fishing, for example, a hooker is distinguished from a netter, and of course, a hooker is a filcher, as explained at length in my post. Consider also hook it “to make oneself scarce,” discussed at some length in my recent dictionary of idioms. As to hockey, yes, the word may be related to hook, but some details are obscure.
Another observation concerns the impact of words related to the sexual sphere on their sources. Obviously, gay “homosexual” is an extension of gay “merry, etc.” The development of the now prevalent sense has been traced in detail, but what matters is that gay “homosexual” killed all the other senses. I once noted that in A. T. Hatto’s splendid English translation of The Lay of the Nibelungen, gay (that is, richly dressed) knights appear with great regularity (this usage was already known to Chaucer), much to the amusement of my undergraduate students. The same happened to queer. In the mid-1950s, I learned and was puzzled by the word because of my lifelong admiration for Oscar Wilde’s works. The phrase snobs queer caused the 1894 trial. Was it possible anywhere in the English-speaking world to say in 1957 “what a queer (= odd, quaint, bizarre) dress,” without making the interlocutor blush or giggle? Many older people can undoubtedly answer this question. The same happened to proud hookers, alas! Today even hooker “ship” is funny. On the other hand, only forty years ago, it sucks was taboo, and now my computer, this precocious child of artificial intelligence, stubbornly suggests three answers to every letter I receive and lists this variant as an expression of my indignant disapproval. Cool!
As for filch (see the post Prolegomena to the word hooker for 19 April 2023), the connection with filchman is obvious, and that is why I referred to Jamieson. The putative Romany etymology of filch is also well-known. But it is far from clear whether the Romany word is not a borrowing of the cant of English thieves. And this brings us to the serious question of alternate hypotheses of word origins. I was asked what I think of the derivation of the adjective secure. Latin had sēcūrus and apparently, sicurus, because the latter was taken over by the speakers of Old English as sicor (Modern English sicker is dialectal, but compare German sicher “certain,” from sihhur). Could the English word be borrowed from a similar-sounding Greek one and be traced to a cognate of Latin sīcut “like” and understood as sic-urus? When there is a secure explanation, why look for another one? How is the vowel length in sīc– to be explained away, and what is –urus? Our correspondent borrowed his etymology from someone who, apparently, had little exposure to historical linguistics. People not versed in etymology often have good ideas about the origin of slang, technical terms, puzzling idioms, and the like, but the history of old words is a special subject, and guesses about their past are seldom, if ever, profitable.
Trawl has traditionally, even if with reservations, been derived from Latin trahere “to draw.” All the senses of troll (“move about; roll; sing in full round voice”) may be related. The difficulty is that tr– and dr– verbs are often either sound-imitative, sound-symbolic, or both (compare trip, drip, trap, trickle, tread, tremble, trample, trill, drill, drawl, droll, drool, and so forth). This means that all of them might emerge in their respective languages independently of one another or influence their look-alikes, or serve as their sources, or be borrowed. Perhaps trawl has been borrowed or coined more than once with vaguely the same meanings. As regards the verb troll, the existence of stroll complicates the picture, and we wonder: is stroll a variant of troll with so-called s-mobile, mentioned in this blog more than once (a ubiquitous prefix without a clear function)? There is no way to answer this question, because we always face possible, even probable, but not definite solutions.
What is tizzy, part the phrase in a tizzy “in a state of agitation”? Given such a funny word as tizzy, it is almost natural to refer to it as an expressive coinage and to stop wondering about its origin. Compare drizzle, sizzle, swizzle “a frothy (!) intoxicating drink,” and fizz ~ fizzle. Once, tizzy also meant “sixpence,” apparently, an alteration of tester, itself an alteration of teston (the same meaning). Tizzy is a word easy to coin (the pun and half-rhyme are not intentional). There once was the word tuzzy-muzzy “a nosegay.” Dialectal tuzzy means “a tuft of hair” and must be related to touseled, which has respectable relatives but seems to be a member of the same group. The fuzzy Tootsie-Wootsie is also close by.
Can step change “a significant change” be an alteration of sea change? Similarly, can thin as a rake be an alteration of the much more transparent phrase thin as a rail? In the absence of recorded intermediate stages, this type of reconstruction leads nowhere. Also, in connection with rake, our correspondent noted that the Hebrew for “rake” is a word spelled with the consonants resh-het-tof. If rake is of sound-imitative or sound-symbolic origin, then the combination r-h ~ h-r looks natural, but, as always, sound imitation is absolutely obvious only in cases like blah-blah and barf–barf.
Sliding down in a handbasket to hell, I agree, is easy to visualize. It is the ascension to heaven by the same vehicle that remains a puzzle. What is the means of transportation? I asked our readers whether anyone could make sense of the phrase five miles to hell, where Peter pitched his waistcoat. The puzzle has not been solved, but one of our correspondents remembered an old man saying out where Jesus lost his sandals. This image makes more sense! Finally, I received a query whether I have any dirty idioms in my book. Alas, no! In the journals I screened for Take my Word for It, almost everything was prim and proper: in those days, smut and profanity were not allowed in print. The samples sent to me are a delight, but I dare not share them with the readership of this blog. In our speech, we are so squeamish, are we not?
Did both ever refer to more objects than two? No. Bo– in both has cognates in and outside Germanic and has the same meaning everywhere. I believe that in the phrase both Father, Son and Holy Ghost through all eternity, Father and Son was taken for one unit. I am also grateful to another reader who pointed out that the word sib is very much alive in the genetic community and means “sibling” in the speech of its practitioners. And one more. A teacher wrote me that explaining a spelling rule by reference to history is often helpful. I agree. We have rite, right, write, and wright. While dealing with an advanced learner who wonders why English has four written images for the same phonetic unit, a glance at the past may be valuable. But the problem is that spelling is taught (or should be taught) at an early age, and we cannot delve into the history of English while dealing with children. In explaining the spelling of Kate, cat, car, and care, history is good for dessert but not as the main dish.
Once again: please send me questions and notes, and I’ll do my best to comment on them in my gleanings.
Featured image by quintinsmith_ip via Wikimedia Commons (CC BY 2.0)
What does a technical writer do?
When people think about careers in writing, they may focus on writing novels or films, imagining themselves as the next Stephen King or Sofia Coppola. They may aspire be a poet like Tracy K. Smith or Ada Limón. They may lean toward non-fiction, aiming to become an author like Jill Lepore or Louis Menand. But for steady work, there is nothing like technical writing for science, medicine, manufacturing, finance, retail, and other specialized fields.
According to the US Bureau of Labor Statistics, there are more than 55,000 technical writing jobs in the US, projected to hit 59,000 by 2031. And while there are advanced certifications available in specialty fields and even master’s degrees in technical communication, technical writing is a career path open to writers with undergraduate degrees in English, communication, linguistics, and related fields. It usually helps though to have some experience in design, business, science, or technology.
Technical writers prepare instruction manuals, guides, technical articles, descriptions, posts, and documents for all manner of processes, products, and procedures. One student of mine working as an intern said that her first task was to write instructions for packing jars of peanut butter in shipping boxes. Another developed a safety manual for a supermarket’s deli kitchen. A third got assigned to explain procedures for using water softeners and filtration systems. And I once worked for a time documenting an automated query system for real estate listings.
When I’ve invited career technical writers to visit my classes, they’ve shared some of the work they do at any given time. They’ve written clinical evaluation reports and protocols in the medical field and instructions for financial and business systems and software. They have written explanations of how environmental data is collected and analyzed and procedures for compliance with federal and state regulations. One rewrote hospital incident reports to make sure they were understandable to staff and insurers. Another developed a design proposal for an event venue. The writing itself is tremendously varied.
A writer friend once told me, “A lot of writing is not writing,” meaning that there was more to it than words on paper. That’s certainly true for technical writing. Research and consultation are large parts of the technical writer’s work. Figuring out the needs of the users takes time, curiosity, and perseverance. Technical writers consult with designers, developers, managers, technical staff, and end-users to understand the products or processes involved. They may be responsible for recommending the most appropriate medium and design for materials and for ensuring that content is uniform across various modes of delivery.
Some technical writers may also be responsible for collecting and analyzing user feedback and usage patterns. And they may work as technical editors for projects developed collaboratively or drafted by others: scientists, clinicians, programmers, and engineers may not be familiar with writing for the general public or for their end-users.
At times, technical writers can also find themselves doing some feature writing in addition to the technical bread and butter. A writer might profile staff members explaining their work and write newsletter, blog, or magazine copy to tell the backstory of a new discovery, promote an innovative product, or explain a live-saving procedure.
Our fiction-writer colleagues like to remind us that “all writing is creative writing.” If you enjoy research and can write clearly, technical writing may be the place to put your creativity to work.
Featured image by Amy Hirschi via Unsplash (public domain)
This is not a mystery tale, but a few twists in the history of the word riddle are worthy of note. Riddles, it should be noted, played and play an outstanding role in world culture. To win a bride, suitors had to offer or solve an insoluble riddle, as happened, for example, to Samson and Turandot. Forever memorable is the riddle of the Sphinx. According to an Old Norse myth, when the shining god Baldr was killed, Odin (that is, Óðinn) whispered something to his dead son on the funeral pyre. For a long time, the best Scandinavian antiquarians have been trying to guess what Odin imparted to his son, a task as profitable as chasing a rainbow or nailing a pudding to a tree (in an earlier post, I referred to a sleeveless errand). Baldr is dead, to begin with, and if Odin wanted the secret message to be known, he would have pronounced it aloud. In principle, all riddles defy solution: they are meant to be insoluble. Collections of riddles from many cultures make such an unexpected conclusion clear. The scholarly literature on this genre of popular culture is a shoreless sea. Especially instructive is the excellent recent book byProfessor Savely Senderovich Riddle of the Riddle.
English has two homonyms: riddle “puzzle” and riddle “sieve” (the latter, I think, mainly northern English). Their origin has been known for centuries and can be found in any middle-sized dictionary, so that I won’t be able to say anything new about them. However, I have a tiny postscript, and it is for its sake I am telling this story. But first, here is a short and trivial note on the origin of riddle1 and riddle2.
The root of riddle “puzzle,” from rǣdels(e), is Old English rǣdan “to read.” Reading, it will be noted, is unlike swallowing or breathing: it is a skill, not an instinct. Literacy came to the Western “barbarians,” including the Anglo-Saxons, from Latin (with the conversion to Christianity), and they had to invent a word for the new occupation. Along that road, each tribe went its own way. Verbs meaning “think, guess; collect; count; carve, scratch,” and so forth were pressed into service. The closest cognates of English read are German raten “to advise,” Dutch raden “to advise; guess,” and Old Icelandic ráða (ð has the value of th in English this) “to rule, advise; explain.” Modern English still has the phrase to read a dream and another, semi-tautological, one, especially instructive in this context, “to read a riddle” (read “interpret”), though today the tautology is lost on us.
German Rätsel “riddle”is still an obvious cognate of raten, and the same holds for Dutch raadsel, related to raden. Such, however, is not the case of English riddle ~ read. Two sound changes severed the ties between the verb and the noun, and nothing connects those two words in the language intuition of today’s English speakers. Since the vowel in read is long, we might expect riddle to rhyme with beadle, but in the noun, that long root vowel was long ago shortened before the suffix. It might remain long, as happened in ladle (related to lade “to load”), bridle (related to braid), and many other similar words, but quantitative phonetic changes are rarely consistent.
Some words succumb to them, while the others manage to stay intact, and it is hard to account for their power of resistance. For example, English silly (with its short vowel) is related to German selig “happy” (with its long vowel preserved), while breeches is spelled with ee but rhymes with riches, rather than reaches. Likewise, sieve looks like a perfect rhyme for receive and deceive, but in fact, its poetic partners are live and give. To make matters even worse, the final consonant s in the old form of riddle was taken for the ending of the plural and dropped. The history of –s in English is a fitting plot for a short story. For example, alms, eaves, and riches were once singulars. Speakers misinterpreted –s in them but let it live, while in riddle, s was dropped, to reemerge only in the form of the plural.
Riddle “sieve” traces to Old English hriddel, related to the synonymous form hrīder and the verb hrīdrian “to sift” (sifts of course exist for sifting). The suffix –el in this riddle is the same as in riddle1, ladle, and the rest. Centuries ago, many English words began with hr-, hl-, hn-, and hw-. This initial h has been lost, except in the pronunciation of those who still distinguish between witch and which (in today’s erratic spelling, the old group hw– is reflected as wh-: consider which, when, who, and so forth). Thus, riddle1 and riddle2 became homonyms because of a series of phonetic changes, a most usual scenario. As a result, a modern proposal can be riddled with riddles to the despair of evaluators and the joy of punsters.
And now the promised desert should be served. There is another English word for “sieve,” namely, temes, going back to tamis “worsted cloth, originally used for straining sauces,” familiar from Indian cooking. Tamis “sieve, colander” is a living Modern French word. It has been known for centuries, but its origin remains disputed, and unambiguous facts do not support the suggestion that the Romance noun is an adaptation of the Germanic one. We will leave its etymology in limbo and only note that temes sounds like Thames. This fact gave rise to the hypothesis that in the popular saying to set the Thames on fire (usually about someone or some event that cannot set the Thames on fire) refers to that unfortunate sieve, rather than the river name.
Perhaps this wild guess would have been dead on arrival if it had not been supported, with a most dubious reference to material culture, by E. Cobham Brewer, the author of the once immensely popular Dictionary of Phrase and Fable. The book was on “everybody’s” desk. For years, people kept discussing whether in the process of sifting, some sieves or strainers may catch fire. Citations in my database go all the way from 1865 to 1921(!). Finally, sieves left the drama unscathed, and the Thames was allowed to remain inflammable. Analogs of phrases about rivers set on fire exist in several languages. The origin of the English idiom has not been found, and the OED has no citations of it before 1720. Such a late date makes one suspect that we are dealing with an adaptation of a similar idiom in a language other than English. Be that as it may, the entire argument looks like a parody of etymological research: some people who have more time on their hands than they need launch a ludicrous hypothesis, and the rest of the world keeps the ball rolling until the last thread disappears in the sand.
Featured image: “The Great Fire of London” by Josepha Jane Battlehooke, Museum of London, via Wikimedia Commons (public domain)
The promise, referred to in the title, was made last week (19 April 2023) in connection with the etymology of the verb filch. I also indicated that my exposition would be written in honor of and as a tribute to Professor E. Peter Maher, who had explored the history of hooker in exhausting detail. Everything I will say below, except for a few remarks in brackets, will be borrowed from his book-length publication “The Unhappy Hookers; Origin of hooker ‘prostitute’,” published as a special issue of the monthly journal Comments on Etymology 50, 6-7, 2021 (58 pages, with numerous illustrations and an exhaustive bibliography).
Maher’s main point is that hooker “prostitute” swept (American) English only after the publication of Xaviera Hollander’s (XH) book The Happy Hooker (1971). Twenty-five million copies were sold, and “everybody” read it. The author’s real name was Vera DeVries. There is an entry on Hollander in Wikipedia, and I’ll dispense with retelling it. XH of course did not coin the word hooker. Occasionally, it even had some connections with sex, but before 1971, it had not yet become common property (far from it). In his youth, Maher did not know hooker “prostitute.” Family names are often incredibly offensive, but the many Hookers faced no opprobrium. For example, the founder of the Connecticut colony was the Rev. Thomas Hooker and Harriet Beecher Stowe had a sister who was a Hooker, Mrs. John Hooker, i.e., Isabella Beecher Hooker.
At one time, hookers made hooks, just as carters made carts. But hookers also hook (players play, hookers hook, etc.). That is why thieves were called hookers. The Century Dictionary (a great American reference book) cites hooker ~ hoker “thief,” with an example from a 1598 dictionary: “A cunning filcher, a craftie hooker”, and Maher quotes Mark Twain: “…while Aunt Polly closed with a happy scriptural flourish, Tom hooked a doughnut,” that is, he “filched” it. Maher continues:
“Equipped with hooks, thieves snatched valuable clothes and bed-clothes through open windows and doors. Poor girls and boys of England and Ireland could be sentenced to transportation to Australia for stealing a handkerchief. [Those who have read Oliver Twist will know how true this statement is.] Shoplifters were termed hookers from their modus operandi.”
Indeed, they behaved like filchers (sorry for harping on the familiar note).
Hooker “a glass of whiskey” is “a word of unknown origin.” Was it not called this for its intoxicating purposes: it “sort of” hooked the drinker, did it not? We also find hooker “a boat for fishing with hook rather than net” (from Dutch). Amusingly, as late as 1979, there was an Irish ferry called The Happy Hooker. “The captain must have heard of Xaviera and hoped to attract American tourists,” Maher noted. Enter General Joseph Hooker (1814-1879). The morals of the servicemen in his army were notoriously low. Yet despite the obvious connection, the word hooker owes nothing to the Union general. Some websites keep clinging to this discredited etymology.
Maher found only one pun on hooker before World War II. Even before the 1960s, hooker appeared very rarely in magazines, and the word never made anyone blush. [By blushing I mean unwanted but inevitable associations. For instance, when we speak about the poop on a ship or see the word nincompoop, we cannot suppress our knowledge of what poop usually means.] Later, the word, though not yet vernacular, gained some popularity, but only in the early seventies, after Xaviera, did “all whoredom break loose.” That the 1901 volume of the OED missed hooker causes little surprise: such “vulgarisms” were not allowed in print and could be smuggled in only in the hope of being overlooked, as happened to the bird name windfucker, but more probably, the OED’s team did not have any convincing citations of the word. Once the censorship of this type was abolished, prohibited nouns and verbs swamped the printed page.
The OED team was of course familiar with John Russell Bartlett’s Dictionary of Americanisms. The post-World War Two Second Supplement to the Oxford English Dictionary, cited hooker, as it cited the F-word, and referred to the relevant passage from Bartlett: “Hooker … strumpet, a sailor’s trull, so called from the number of houses of ill-fame frequented by sailors at the Hook (i.e., Corlear’s Hook) in the city of New York.” Maher remarks that Bartlett’s hooker indeed designated someone living or frequenting Corlear’s Hook in New York. [It was a word like Londoner or New Yorker. There is a village in Buckinghamshire, England, called Wing. I assume that its inhabitants are called Wingers. Some of them may be wingers.] Corlear Hook’s hookers were prostitutes, but that local name is not the beginning of the now universally known word. Bartlett discarded his entry on hookers as sailors’ trulls from the fourth and the fifth edition of his dictionary.
Though it is true that in the works of Hemingway and his contemporaries, hooker, denoting a woman who could be easily hooked, does occur at irregular intervals, those were occasional coinages. They were easy to understand, but their occurrences do not show that English had an accepted noun hooker “prostitute.” In 1845, a young man tells a friend that “he will find any number of pretty Hookers in the Brick row not far from French’s Hotel,” who do it for love (!). The story sounds almost too good to be true, but those selfless Hookers were obviously not prostitutes. In any case, they were not the immediate ancestors of Xaviera’s happy hookers. Maher’s ironic comment is: “Prostitutes hook a john. Honest women hook a husband.”
Hooker “prostitute” also turns up in later American English, and Maher quotes a passage from a tale told by an Englishman, a naïve outsider but not an idiot, about an encounter with a woman in 1863, whom, from his description, his interlocutors identified as a “hooker,” that is, a prostitute. Quite a few later occurrences make this sense of hooker certain. More examples follow that show the same: hooker “prostitute” existed, but though people understood it (the context was unambiguous and could not be misinterpreted), it remained a rarity.
“The Unhappy Hookers…” is a riveting personal memoir by a seasoned linguist who did not set out to “prove” anything. His aim was to dispel etymological myths by investigating the history of one word, and he did it in an exemplary manner (but I in general have a soft spot for monographs devoted to a single word, be it ginger, shyster, kibosh, or hooker). I’ll now quote a postscript, apparently, added in 2021:
“As I write this I’m a geezer though still ambulatory. My Cold War army hitch was over fifty years ago now, but it seems like yesterday. I can still see it, hear it, smell it, taste it. None of us in the ranks called a prostitute a hooker then. Novelists Wolfe and DosPassos and Steinbeck, born between 1898 and 1902, were now in early middle age. They were writing their novels in the twilight of the Civil War veterans. Some sixty years after the end of the Civil War Bonnie and Clyde were on the road. I was a baby. In this span of time the H-word spawned and spread….”
The final picture on p. 49 is of Bonnie and Clyde.
Featured image by Anthony Mac Donnacha via Wikimedia Commons (CC BY-SA 3.0)
Prolegomena to the word <em>hooker</em>: the English verb <em>filch</em>
My initial idea was to write a blog post about the history of the word hooker, in order to celebrate the extensive research into this question by Professor J. Peter Maher, and I’ll do so next week, but it suddenly occurred to me that as regards meaning, the verb hook has some affinity with the verb filch, and I decided to write what I know about it. One can find an entry on filch in An Analytic Dictionary of English Etymology: An Introduction,published fifteen years ago (2008). There is a brief mention of my hypothesis in Etymoline, but without reference to the source or even minimal discussion. Nor did the omnivorous and omniscient Internet take cognizance of what is said in An Analytic Dictionary about filch. Below, I’ll make the idea of that entry common property, assuming that the hungry readership will pound on this post.
Problems emerge the moment we begin to explore the history of filch, because two homonymous verbs exist: filch “to attack” and filch “to steal.” They are almost certainly unrelated. The earliest attestation of filch “to steal” goes back to the second half of the sixteenth century, and, most probably, the verb entered English at around that time. Therefore, searching for its Old English ancestor looks like a sleeveless errand (see the post for 26 April 2017, on this curious phrase). The first book-length dictionary of English etymology was published in 1617, so that its author (John Minsheu) knew filch, and indeed, he offered a hypothesis on its derivation. All his followers also wrestled with filch, and now, four centuries later, responsible sources have a solid answer to the problem: “Origin unknown.”
Those interested in etymology are well aware of the fact that in dealing with slang (and filch is of course slang), historical linguists seldom succeed in finding such words’ unquestionable sources. The same holds for filch (hence the shocking verdict “origin unknown”), but tracing the pathfinders’ way to the truth, even if it is a devious way with no dazzling light at the end, cannot be called a futile endeavor. At the very least, it prevents later researchers from repeating the mistakes of their predecessors. I’ll briefly go over the attempts to discover the etymon of filch, but without giving references. Those interested in the bibliography will find all they need in my dictionary.
For quite some time, Greek fēlós “deceitful” and Latin fallax “deceitful; fallacious” looked like the desired etymon, but filch was probably coined or borrowed by those English speakers who knew more about filching than about Greek and Latin. Besides, filch is a verb, rather than an adjective! Old Portuguese filhar “to seize,” perhaps allied to Italian pigliare “to seize” or French piler “to crush,” have also been pressed into service by ingenious word hunters. Thieves’ secret lingo has always been to a certain extent international, because at all times, criminals used to cross the borders of their native regions and made attempts to communicate with one another, ideally without being understood by the uninitiated. Their cant (as this language is called) has been explored by several excellent scholars. The Old Portuguese verb does look like a fairly good fit, but the source is suspicious: why Portuguese, and why no modern synonyms elsewhere in the Romance-speaking world?
Germanic words related to filch seem to hold out better promise. Therefore, filch has been compared with Old Icelandic fela “to hide” (however, the senses do not match!) and especially with Gothic filhan “to steal.” Apart from the -h ~ -ch mismatch, the latter verb looks like a feasible cognate of filch. Gothic, it will be remembered, is an Old Germanic language, known mainly from a fourth-century translation of part of the New Testament. Filhan may have a more secure cognate in English, namely, the regional verb feal “to hide,” though it should be repeated that “steal” and “hide” are not synonyms. Curiously, Stephen Skinner, the author of a 1671 etymological dictionary of English (the second ever published), seems to have cited a reasonable cognate of filch, namely, German filzig “feltlike,” with numerous figurative senses, from “sleazy” to “corrupt; greedy,” but no one seems to have taken his suggestion seriously. Filz is related to English felt, the name of a fabric.
Many old etymologists were content with stringing together various look-alikes, without even wondering how filch could develop from or be related to them. Dutch fielt “rascal,” Swiss German flöke “to steal,” French filouter “to cheat” (French filou “crook, rogue”). Irish Gaelic fealleaidch “knavish,” Old English flōcan “to beat, strike,” and quite a few others have been compared with filch. Filching presupposes a sleight of hands, and it may not be due to chance that in so many languages, words designating all kinds of quick and unsteady movement begin with the sound–symbolic group fl. Compare fly, flow, fleet, flimsy, and the like. However, spotting a vaguely similar word, without showing how it could reach English and yield filch, is a wild goose chase.
The consonant ch did not exist in the earliest form of English. Theoretically, filch can go back to some form like filk. Norwegian and Scottish pilk(e) ~ pilk mean “to pick,” and it has been suggested that filch is a rhyming synonym of pilk. This derivation is not impossible but rather improbable. As often mentioned in this blog, the more ingenious or convoluted an etymology is, the smaller the chance of its providing a clue to the riddle. I think the same also holds for the comparison between English filch and Danish regional filke “to scrape.” And we are facing the familiar question: how did the Danish dialectal verb become known in English? In my opinion, filke as the etymon filk looks as improbable as pilk. The answer should be closer to home.
The great Scottish lexicographer John Jamieson recorded filchans “rags, patched or fastened together, hooked in a bundle,” and we can return to Skinner’s idea. German verb filzen means “to comb through,” and in Middle High German (that is, in a language spoken many centuries ago), filzen already meant “to search a person” (from Filzer “comb”?). The inspiration for all such words was the name of the fabric felt (German Felz). An Early English filcher ~ filchman was, apparently, a person who could use his hook for stealing things. Let us note that English filch was recorded late, and filk, its putative base, has never been found. Most probably, it never existed. As far as we can judge, filch has always meant only “to pilfer.” Filch “hook” seems to be the best starting point for tracing the origin of filch “steal.” Also, hook is a familiar starting point for all kinds of slangy phrases. Compare hook it! “make yourself scarce,” hook up, and of course hooker, the subject of next week’s post.
Filch, it appears, is an English adaptation of German filzen,a term of the thieves’ (international) slang. The German verb ends in z, that is, ts, which, obviously, could not survive in English and was replaced with ch. In principle, I am returning to the suggestion first made in 1671. Three and a half centuries of fruitless wanderings? No, not quite fruitless. Rejecting implausible guesses is part of all good research. Hence the inevitable sleeveless errands and wild goose chases. Scholarship exacts a high price, but all is well that ends well.
A rake’s etymological progress to hell
Another punning title! I am so sorry. This post is about a garden instrument, not the profligate, immortalized by William Hogarth and Igor Stravinsky. Last week, I wrote about the verb scratch and promised to discuss the origin of rake. The superficial factor that unites the two words in this blog is the letter in which a reader asked me about their derivation, but as will be seen, there is a deeper link between them. And hell will turn up below more than once, with my good intentions taken for granted.
Three English words sound as rake: the two mentioned above and rake “inclination from the perpendicular” (a sailing term). Though at first sight, they do not seem to be connected, I’ll try to show that their histories perhaps intertwine. Rake “implement” is related to Old Norse reka and German Rechen, a southern word (from rehho). This Rechen has a northern synonym, namely Harke, attested first only in the sixteenth century. Its origin is murky, but rehho and Harke do sound somewhat alike. Rake “profligate” seems to go back to rakehell, that is, a person for whom hell is raked (prepared, made available?), such a compound as puzzles us until we learn its history (think of spendthrift or cocktail).
The sailing term rake is one of several seventeenth-century nautical words of obscure antecedents (like awning and tarpaulin, for example; both have been discussed in this blog). Rake “implement” is related to the verbs meaning “to reach out, stretch, remove,” and it resembles English reach, from raikjan. Then we notice German ragen “to protrude,” most often used with some prefix. A few related forms in Scandinavian also exist. Judging by its Old English cognate, rake, the name of a gardening implement, once began with hr-. And to repeat, an etymologically mysterious synonym (doublet?) of German Rechen “implement” is called Harke. We seem to be dealing with two competing sound groups—rk and hrk—for “rake” and the other words being discussed here. For Harke a sound-imitative origin has been suggested, though cautiously. The others have been traced to the basic meaning “to reach out.”
Writing a blog is different from writing a scholarly article. In a post, one may risk an unsafe hypothesis, listen to a refutation, and, if necessary, recant. I would therefore like to suggest that the initial impulse behind the oldest words mentioned above was to render the noise produced by scratching (this is the link to the scratch–kratzen pair of last week). They came to designate all kinds of objects, including a few protruding ones. After all, a rake is a big, long thing. Reaching out, an angle at which the prow rises above the surface, and so forth, seem, I suspect, to be products of secondary associations.
Perhaps the idiom thin as a rake will support my idea. William. B. Lockwood, mainly known for his studies of the origin of bird names, occupied himself one day with raking together the cuttings on the lawn, and a thought occurred to him that there is nothing thin about a rake. (Remember an apple falling on Newton’s head? All dedicated researchers are alike: their problems never leave them in peace.) No one, Lockwood added, says: “thin as a hoe.” Struck by this idea (and it is surprising that it had not occurred to anyone before him), he decided to derive the word in the English idiom from the Old Icelandic root hrak– “wretched, wicked.” I agree that thinness is not a distinctive feature of a rake, but apart from some phonetic difficulties (which I will ignore here), I wonder whether the Scandinavian noun was known widely enough in English to become part of the simile. Lockwood’s supporting material is meager. Also, it might be good to find a similar idiom in a Scandinavian language, but it does not seem to exist anywhere outside English.
Lockwood’s idea, first published in 1973, never lost its appeal to him, and he reproduced it in his 1995 book An Informal Introduction to English Etymology. Philip Durkin, Deputy Chief Editor and the chief etymologist of the Oxford English Dictionary, shared Lockwood’s doubts, but in a 2012 paper, referred to a fact of material culture: “… the implement used by medieval farmers appears to have been rather more similar to the modern soil rake… …perhaps the expression arose from the similarity of a soil rake to the skeletal form of an emaciated figure….” Since the simile was already known to Chaucer, Durkin’s idea looks plausible, and, in an indirect way, it may confirm my suggestion that the old rake, a kind of hoe, did not protrude or stretch, or reach out, but mainly cut-cut-cut, perhaps producing hrk-rk sounds. As Durkin puts it, opinions are likely to continue to differ. But if my hypothesis has any merit, it follows that the names of the objects mentioned above did not go to an ancient root meaning “to heap up, reach out; arrange” or something similar. Such a root may not even have existed.
In an old post, I once told a story of a little girl asking her mother about how the streetcar (tram) functions. The woman explained everything in a most scholarly way. The girl listened patiently and then said: “No, this is not the way it goes.” “And how does it go?” “Ding-dong.” I suspect that the ancient rake went hrk–rk or rk–hrk. Those who will take my musings seriously may also consult the posts on trash and rubbish (24 and 31 March 2021), and those who are interested in the origin of other implements are invited to read the posts for 25 March 2020 (adze) and 14 September 2022 (mattock). Mattocks and rakes belong together, at least to a certain extent.
“In Sussex the devil is called ‘the poor man’, and the earthworks of the ancient British hill-fortress at the well-known Devil’s Dyke (near Brighton) are known as ‘the Poor Man’s Wall’.” The great Max Müller responded to the author of this note and pointed out that the extraordinary features of nature and even works of art are ascribed to the devil all over the world, for which reason the philologist Herbert Coleridge called the Devil Pontifex Maximus, a pun with an obvious reference to Latin pons, pontis “bridge.” Does anyone know the origin of the phrase poor devil (French pauvre diable, German armer Teufel)? Who coined it and when? Neither thin as rake nor poor devil occurs in my rather complete database featuring the etymology of English idioms. It means that for three centuries no one has discussed them in popular periodicals or scholarly journals.
“The Devil’s Mouth. The term was applied by the Spaniards to the Old Mole battery, whose fire caused them much annoyance when they were besieging the fortress, and the Devil’s Tongue is the point or spit of ground on which the Old Mole battery is built. The Devil’s Tusk is a pinnacle of limestone in the shape of a tooth, and about thirty feet in height, at the rear of the Royal Naval Hospital” [the author did not specify which hospital is meant].
TOPOGRAPHIA INFERNALIS is endless, because the Devil is a famous rake. End of pontificating.
Featured image: William Hogarth, The Tavern scene from “A Rake’s Progress”, Sir John Soane’s Museum, via Wikimedia Commons (public domain)
My son’s language is made of a bundle of sounds that do not exist in the Spanish that we speak around the Río de la Plata. He repeats syllables he himself invented, he alternates them with onomatopoeias, guttural sounds, and high-pitched shouts. It is an expressive, singing language. I wrote this on Twitter at 6:30 in the morning on a Thursday because Galileo woke me up at 5:30. He does this, madruga (there is no word for “madrugar”, “waking up early in the morning” in English, I want to know why). As I look after him, I open a Word document in my computer. I write a little while I hear “aiuuuh shíii shíiii prrrrrr boio boio seeehhh” and then some whispers, all this accompanied with his rhythmic stimming of patting himself on the chest or drumming on the walls and tables around the house.
My life with Gali goes by like this, between scenes like this one and the passionate kisses and hugs he gives me. This morning everything else is quiet. He brings me an apple for me to cut it for him in four segments. He likes the skin and gnaws the rest, leaving pieces of apples with his bitemarks all around the house. He also brings me a box of rice cookies he doesn’t know how to open. Then he eats them jumping on my bed. He leaves a trace of crumbles. Galileo inhabits the world by leaving evidence of his existence, of his habits, of his way of being in the world.
When we started walking the uncertain road to diagnosis, someone next of kin who is a children’s psychologist with a sort of specialisation in autism informally assessed him. She ruled (diagnosed, prognosed) that he wasn’t autistic, that we shouldn’t ask for the official disability certificate (because “labels” are wrong, she held), and that he should go on Lacanian therapy and music therapy on Zoom —now I think this is a ready-made sentence she just gives in general to anyone.
The most violent intervention in Galileo’s subjectivity is denying his being-disabled in an ableist world and his being-autistic in an allistic world. We, as a culture, have internalised the terror of disability so deep in our minds that we hurry to deny it. We are not willing to accept that what causes us so much angst and dread actually exists, that it is not an imagined ghost. Denying like this, in this delusional way, is an instinct only humans have. It is so human (so stupid) that it is not a survival instinct. When we deny autistic affirmation, we prepare the ground for its annihilation, i.e., for the annihilation of everyone who is autistic. Being autistic isn’t being an imperfect allistic, a not-yet-allistic person. Being disabled isn’t the same as being a flawed abled person. The denial of disability doesn’t amount to affirming an alternative ability, it implies the ableist annihilation of all vulnerability. But when disability is negated, able people do not survive either. We are born and we die in disability. How did it happen that we dare to imagine we can supersede need? (Maybe by the same process by which it is believed that capitalist profits are meant to satisfy human needs).
The instinct of denying disability is not innate, though. It is an intelligent trap designed to break communities apart, to disorganise, to debilitate us: not to make us disabled but to make us unable, powerless. This is how ableism works, de-politicising vulnerability and unease, making disability, at most, an object of pity and compassion, a matter of bad luck, a fate to try to twist and avoid.
***
Galileo’s spoken language has the musical texture of a genre he alone can perform. There was a (short) time when I thought that my role in his life was to be her translator, a mediation between him and the rest of the world. This is impossible for many reasons. The most important of them isn’t that I don’t get him (I don’t), it isn’t that I don’t speak his language (I don’t), or that no one (much less a mother) can or should mediate anyone. The main reason is that Galileo speaks as someone who plays in their instrument a piece that they have composed for themselves.
Sometimes language is comprehensible only insofar as one gets ready to listen to it as if they were in an empty church in front of a little bench where Rostropovich is about to play Bach’s suites with his Duport, and as if he were Bach himself. Then, and only then, we understand that we don’t understand, that we are at the gates of the incomprehensible. When is language more language than when it is spoken so incomprehensibly? The impossibility of interpreting oneself, myself, comes not only from the fact that no one controls or owns language. No one plays their own scores because no one creates their own language. No one, but Galileo and his equals. The autistic non-verbal language is that impossible thing that we try not to talk about when we talk, that we try to drown by talking too much, moving our hands, and writing for example this text. Autistic languages say what can’t be said in any articulated allistic “normal” language. Galileo speaks a language that complements other languages. This language of his is not the opposite of language: it perfects other languages, like music or silence do.
***
Does my son have a mother tongue? Do we speak to each other as mother and child? What do we tell each other when we chat? My son’s autism and his magic words lend me a whole new vocabulary for my own neurodiversity, a new and authentic view on my severe misophonia, hyperacusis, and hyperosmia, and on my life-long inability to grasp the majority of the rules of interpersonal relationships, among other things I thought were personal flaws that made me inferior. I won’t mask it anymore. I won’t keep it a secret anymore. Now I know how to talk about it, now I have names to name it. Maybe he will never speak his mother tongue or any other “normal” language, but he has taught me to speak a language in which I now can say what I couldn’t formulate in an allistic alien tongue. Stripping me of all the allistic and ableist expectations that have shaped the way I was meant to raise my children has liberated me from the suffering of trying to meet them myself. The truly difficult thing, besides raising an autistic child in an allistic world, besides being a non-verbal autistic child in an ableist world, is how to de-internalise all this life-long inferiorisation.
But I know he will tell me how.
Scratching all the way to hell (second series)
Quite some time ago, I was asked about the origin of the verbs scratch and rake. At first sight, the etymology of both looks unproblematic, but modern researchers are cautious people and see reefs where the rest of the speaking community glides happily over the surface. Today, I’ll confine myself to scratch (oh, how hard it was to resist the temptation of saying that I’ll begin from scratch!).
Scratch turned up in English texts only in the fifteenth century. As usual, it does not follow that this verb was coined so late. My doubts are based on the fact that German kratzen (the same meaning) goes back to the oldest texts in that language. Middle English had two synonyms: skratten and cracchen. Cracchen looks French, and William W. Skeat, our greatest but not infallible authority for the origin of English words, believed that scratch and cracchen had been confused, that is, got into each other’s way and produced some sort of semantic blend. Perhaps they did. It is sometimes hard to tell apart even not fully identical twins.
I notice that Swedish skrata means “to make a lot of noise, to laugh uproariously,” while sratta means “magpie.” Both words are rather obviously sound-imitative. To add to this list, Old Icelandic skrati meant “ghost” (a noisy creature?), and its near-doublet skratti was a designation of the Devil; hence English Old Scratch “Devil.” (See some musings on such words in the post for 2 May 2012 “Further Adventures of Scr-Words.”) Scratching does not produce a lot of noise, but the group scr- ~ skr- does occur in quite a few sound-imitative and sound-symbolic words. Scream and screech are ideal examples. Since a bird called “screech-owl” exists, I may mention Swedish skrata “owl” and skratta “to laugh.” (The Germanic verbs for “laugh” were all onomatopoeic: laugh goes back to hlah-; compare hlah– with cluck-cluck and giggle.) English scrape, a verb of Scandinavian origin (Old Icelandic skrapa), makes one think of a jarring sound. Likewise, English scramble may or may not be a loan from Scandinavian, but borrowed or native, it does not refer to peace and quiet.
In dealing with scr– ~ skr– words, one never knows where to look: everything is so suggestive. Is the English adjective scrumptious indeed an emphatic alteration of sumptuous, as some of our sources suggest? This hypothesis is not entirely without merit. Scrimmage is, apparently, a variant or a close relative of skirmish. On the same note: where is scrouge “to crowd out” from? Dickens’s Scrooge, who scrimped and saved, knew well how to screw people and leave them screwed up. The origin of scrimp is also obscure, as it should be. A doublet of skimp? Even Latin scrībere “to write” (consider scribe, script, Scriptures, and others) belongs to our sound-imitative corpus, because writing began as scratching.
In any case, scratch and German kratzen seem to belong together, despite the gap in the chronology of their earliest attestations. To my mind, it may not be too bold to assume that even in the oldest period, a pair of synonyms sounding like kratzen ~ scratch coexisted. I even wonder whether the s-form is the later one. When we encounter the enigmatic s-mobile, the ancient prefix that attaches itself like a barnacle to the roots of so many words, we assume that the s-less form is the older one. But perhaps (just perhaps), given the coexistence of a great number of words with and without that spurious prefix, some forms with s- are at least as old as those without it. Kratzen is a verb of murky origin, but if we assume that it began its existence with s and then lost it, rather than that scratch acquired its s later (and developed from cratch to scratch), we may solve the puzzle of their relationship. Be that as it may, we are certainly dealing with sound-imitative and/or sound-symbolic formations.
Last week, I promised to continue my series on TOPOGRAPHIA INFERNALIS. Since I did mention the Devil in today’s post, this subject will come in handy. (All the passages below are from Notes and Queries for 1884-85.)
Five miles beyond hell, where Peter pitched his waistcoat. No one could explain the allusion, and the modern Internet provides no help either. And here is another question that received no answer in the eighteen-eighties: “Is anything known of the origin of the saying ‘Elephant-end,’ which place is supposed to be situated ‘where the devil can’t get for nettles’”? Nettles and thistles are the proverbial food of the Devil. Apparently, the rush belong to his diet too (see below). The Devil, unable to get nourishment, is of course hungry and irate. But why elephant-end,and what does the phrase mean?
“‘An eminent book collector, noted for his good nature, declared that a man who published a book without an index ought to be put into thistles beyond hell, where the devil could not get at him’ (Temple Bar, October 1882, p. 191.) Yet the thistle appears to be in some sense one of the devil’s plants. ‘Having met the Lord one day, the devil asked for oats and buckwheat as his reward for having taken part in the creation of the world. The request was granted, whereupon the devil began to dance for joy. The wolf came up and suddenly asked the meaning of this frivolity. In his confusion, the devil forgot what had been given him, and replied that he was dancing for joy at having received the rush and the thistle, to which plants he still adheres”
(Athenæum, Sept. 23, 1882, review of La Mythologie des Plantes, by Angelo de Gubernatis.) [Temple Bar, 1860-1906, was an excellent literary magazine, and so was Athenæum, 1828-1926. I mainly used the latter for book reviews and letters to the editor, while compiling my bibliography of English etymology and my recently published dictionary of idioms.]
Gudbrand Vigfusson (an anglicized spelling of the name of a great Icelandic lexicographer and philologist) published, in English, the following note in connection with Hekla, the volcano about which I wrote a few lines last week: “In an old Danish hymn or song of the sixteenth century I remember having read of a drove or hunt of condemned souls on the way to Mount Hecla [sic] from Denmark. Satan, the drover, called Lureman [from Danish lure “to lurk”?], sings out: “Come! come! come! you must [that is, you must go] to Heckenfield, to Hecken, to Heckenfield, with the swarm of souls into the black hole…” (The Academy, February 14, 1885). [The Academy, 1869-1902, was another first-rate magazine. One sometimes wonders how in the period 1850-1920 the literate British public could support so many popular journals and consume those thousands of pages every week, and there were many, very many more than the three mentioned here. The Wild Hunt was a story known all over Europe. At one time, in Scandinavia, the god Odin was associated with it.]
In a comment on the previous post, our correspondent wrote that he knew the phrase to go to hell in a handcart (rather than in a handbasket). I think both mean the same, that is, “to go from bad to worse,” and in both, the way of propelling the vehicle remains obscure. Perhaps the driver is Old Scratch.
FreshRSS 