In continuation of the theme that you can’t trust any of today’s English dictionaries to reliably summarize what’s known about the etymology of a word, consider how seven dictionaries summarize the etymology of RACQUET as in tennis racquet:

Webster’s New World @ YourDictionary.com: MFr raquette, earlier rachette, palm of the hand < ML rasceta (manus), palm (of the hand) < Ar rāḥa(t), palm of the hand
American Heritage @ YourDictionary.com: Origin: Middle English raket, a kind of handball, from Old French rachette, palm of the hand, racket, from Medieval Latin rascheta, palm, from Arabic rāḥat (al-yad), palm (of the hand), bound form of rāḥa; see rḥ in Semitic roots.
Collins English @ Dictionary.com: Century16: from French raquette , from Arabic rāhat palm of the hand
Merriam-Webster @ M-W.com: Middle French raquette, ultimately from Medieval Latin rasceta wrist, carpus, modification of Arabic rusgh wrist. First Known Use in English circa 1520.
Chambers Dict @ ChambersHarrap.co.uk: 16c: from French raquette, from Arabic rahat palm of the hand.
Concise OED @ Oxford Dictionaries.com: late Middle English: from French raquette, via Italian from Arabic rāḥa, rāḥat- ‘palm of the hand’
Random House @ Dictionary.com: 1490–1500; < MF raquette, rachette, perh. < Ar rāḥet, var. of rāḥah palm of the hand

A symptom that something might be amiss: Merriam-Webster derives racquet from Arabic ‘’rusgh’’ = “wrist”, while the others derive it from Arabic ‘’raha(t)’’ = “palm of the hand”.

CNRTL.fr says the word racquet is first attested in French in 1314 in a surgery textbook, spelled “rachete” and meant the wrist and/or the bones of the wrist. It goes on to say, with my paraphrasing:

Borrowed from Arabic ‘’rāḥa’’ = “palm of the hand” through the intermediation of medieval Latin medical texts. ‘’Rasceta manus’’ = “the carpus”, i.e. the wrist bones, is attested in the well-known 11th century Latin medical writer Constantinus Africanus. “Rasca’’ = “the tarsus”, i.e. the bones of the ankle+heel of the foot, is also attested in 11th century Latin. The spelling ‘’raseta’’ = “the carpus” is in Latin around the same time as the first appearance in French.

Marcel Devic’s book, dated 1876, Dictionnaire Étymologique Des Mots Français D’Origine Orientale: arabe, persan, turc, hébreu, malais has the interesting information that in an Arabic-to-Latin translation of a medical book in the later 12th century, translated by the well-known Gerard of Cremona, the Arabic رسغ ‘’rosgh’’ = “wrist, carpus” was translated to Latin as “rascete”.

Let me say the obvious: (a) the Arabic ‘’rāḥa’’ (and its admissible variant ‘’rāḥat’’) = “palm of the hand” is not close phonetically to the medieval Latin rasceta | rasca | rascete | raseta ; and (b) it is not close semantically. The semantic mismatch is the bigger problem. In the two earliest attestions in Latin, one says ‘’rasca’’ means the bones of the ankle, and the other does not strictly say ‘’rasceta’’ is the bones of the wrist—rather it says ‘’rasceta manus’’ is the bones of the wrist (where manus is Latin for hand, of course). There is no evidence that the Latin word was arrived at from Arabic “palm of the hand”!!

I stand by that last sentence, exclamations and all, but actually there is one piece of indirect circumstantial evidence, and it’s a legitimate piece of evidence, but I argue it’s not enough to warrant the certainty that the dictionaries are communicating about the word’s origin. Constantinus Africanus’s native language was Arabic. They called him “Africanus” because he came from Africa. In his later years he lived in southern Italy and wrote medical stuff in Latin that drew heavily on Arabic medical sources. He probably was born Christian in Tunisia, but possibly was an adult convert to Christianity in southern Italy.

Constantinus Africanus wrote in the later 11th century and died 1087. When CNRTL.fr says ‘’rasca’’ = “the tarsal bones” is attested in another, separate 11th century source (--> xie s., Id. ds Nov. gloss., s.v. navicula <--) I’d like to know who that source was and whether it was independent of Constantinus Africanus. Because if it was independent and antecedent, or just arguably or possibly so, then the Arabic origin story for racquet becomes highly uncertain. If it be allowed that Constantinus is the true originator of the medieval Latin word, an Arabic origin in ‘’rāha’’ or ‘’rosgh’’ must remain uncertain because of the weaknesses you’ve seen.

The 2005 book “Word Origins” by John Ayto joins in the unanimity and confidence of the other dictionaries and summarizes it this way: “Racket for playing tennis [16th century English] and racket ‘noise’ [16th century English] are unrelated words. The former was borrowed from French raquette, which originally meant ‘palm of the hand’. This goes back via Italian racchetta to Arabic rāhat, a variant of rāha ‘palm of the hand’. The origins of racket ‘noise’ are not known....” On the evidence, what he ought to say is: “The origins of racket ‘noise’ are not known.”—I believe it—“The origins of the racket for playing tennis are in medieval Latin medical texts where it usually meant the bones of the wrist. The source of the Latin is uncertain.”

Incidentally and although it’s beside the point, Chaucer’s poem “Troilus and Criseyde”, dating from the 1380s, contains the line: “But canstow pleyen raket, to and fro” [canst thou play racquetball]. There’s no information in Chaucer about the definition of the game “raket”. I’ve seen it said that it was probably a handball game with no racquets (see e.g. the American Heritage Dictionary quoted above). Of the five dictionaries quoted above that state the date of entry of the word into English, all five give a date that’s more than a century after Chaucer. I expect that’s because Chaucer’s word was a game, not a bat. Another point perhaps worth noting is that the French word ‘’paume’’ means the palm of the hand, but it is attested in 1355 and 1373 meaning a game played with the palms of the hand, and today “le jeu de paume” is still a name for tennis in French.

“But canstow pleyen raket, to and fro” [canst thou play racquetball].—Geoffrey Chaucer, 1380s

Tangentially, I won’t accept the report that Chaucer’s “raket” game was a handball game, not a racquet game, unless and until I’ve seen good evidence for it. The medieval Latin word meant the bones of the wrist (and did not mean the palm of the hand). Racquet as in bat comes by extension from the bones of the wrist, I believe.

In continuation of the theme that you can’t trust any of today’s English dictionaries to reliably summarize what’s known about the etymology of a word

I really wish you would stop making statements like this. For one thing, it’s not true. There are English dictionaries (the OED for one) that give exactly the arguments you put forth. For another, it’s cherry-picking. You cite a handful of cases, out of thousands, where some dictionaries don’t accurately reflect the entirety of the scholarship and then damn all the etymologies as incorrect. You make good arguments without needing to resort to such hyperbole.

In the case of racket, the OED makes essentially the same argument as you put forth here. Although it is likely the English word comes from the Middle French racquette, a paddle used for scraping the bottom of a ship, which is attested to as early as 1388. Where the Middle French word comes from is the mystery.

And regarding Chaucer’s use of the word, no one knows exactly what game he was referring to. (Evidently Lydgate mistook Chaucer’s use to mean a game of dice.) The word does not appear to have been in common use in England until much later, so these early uses are lacking context.

In continuation of the theme that you can’t trust any of today’s English dictionaries to reliably summarize what’s known about the etymology of a word,

What are you trying to do - prove that this site is based on a false premise and that there is no reliable etymological resource besides you?  I agree with Dave - please stop basing your posts on a rant.  Your posts are interesting but for me, they are spoiled by bias and hyperbole.

There is an extraordinarily well-documented 9-page article by Christian Schmitt ”Die Araber und der Tennissport” (Arabs and the sport of Tennis) in a book of papers entitled Romania Arabica (1996) which addresses in detail all of the points raised in this post. 

Apart from the linquistic evidence, Schmitt also discusses evidence relating to the history of racket sports, most of which comes from Carl Diem, Weltgeschichte des Sports und der Leibeserziehung.  The “standard” explanations given in French, Italian, Spanish and English etymological dictionaries are reviewed in detail.

Schmitt’s conclusion is as follows:

“. . . lassen es geboten erscheinen, fur frz. raquette “Tennisschläger” eine pikardische Nachfolgeform von Lt. re-captare/*captitare als Ausgangspunkt zu postulieren, das Nomen bildet dabei eine deverbal Rückbuldung . . . Überflüssig sein dürften auch weitere Hinweise zur gebotenen Trennung von frz. raquette “Tennisschläger” von frz. raquette “paume de la main” oder mrfz. rassette, die als medizinische Fachtermini mit dem Arabischen verbunden bleiben. Raquette “Tennischläger” aber ist eine sprachtliche Bildung, die in dem Land entstanden ist, wo nach unserer Kenntnis der Tennissport seinen Ausgang nahm: den nördischen Frankreich.”

roughly:  it necessarily appears that French raquette is a Picardian (northern French) derivation whose starting point can be postulated as Latin recaptare/*capitare, the noun (raquette) then being a back-formation from the verb . . . From this it follows that French raquette ("tennis racket") must be separated from French raquette ("palm of the hand") or Middle French rassette, which as medical terms are still to be considered linked to Arabic.  Raquette ("tennis racket"), however, is a linguistic formation which occurred in the region where, according to what we know, tennis has its origin: northern France. 

All in all a very interesting and persuasive article and highly recommended to those who read German.

What Dave and Eliza said.

Dave Wilton - 21 November 2010 08:07 PM

I really wish you would stop making statements like this. For one thing, it’s not true. There are English dictionaries (the OED for one) that give exactly the arguments you put forth. For another, it’s cherry-picking. You cite a handful of cases, out of thousands, where some dictionaries don’t accurately reflect the entirety of the scholarship and then damn all the etymologies as incorrect.

The OED is different. I’m not knocking the OED. I’m talking about the summaries in the others. All of the words I’ve picked are listed on the Wikipedia page List of Arabic loanwords in English. That page lists either 131 or 164 loanwords, depending on how you count them—around a couple of dozen words are listed as “perhaps” from Arabic. From that list I’ve cherry picked CALIBER, CORK, ALIZARIN, GENET, ALMANAC, LILAC, HAZARD, MASSAGE, RACQUET… and I expect to pull a few more from that page, just a few, and when I find them I’ll post them. So that’s 5% to 10% of the reported Arabic loanwords in English. Close to 10%. I think I’ve shown that all seven of the print-published English dictionaries that are also online-published tend to have a “herd” mentality and all of them are too credulous and uncritical in their reports of those particular words —though some one or another of them says “perhaps” in one or another individual case, they all communicate certainty for most of those words.

The word ALIZARIN is a superb example of what I’m talking about. Since ALIZARIN was buried with the CALIBER thread and went unnoticed within that thread except for the comment that it would’ve been better to have started a new thread for it, here again is the link to the ALIZARIN cherry (or raspberry).
Here’s the link to the GENET raspberry.
Here’s the link to the ALMANAC raspberry.
The CORK raspberry is also under the CALIBER thread but is spread out over several posts with multiple posters.

Well, part of the issue is relying on Wikipedia. The OED lists 597 entries with “Arabic” in the etymology. (Not all of these are from Arabic, some are cognates, etc., but there are clearly more than Wikipedia lists. That would drive your numbers down considerably.

The second is that you are not saying that a select list of dictionaries that you have consulted are at fault. You say that all English dictionaries are unreliable, and furthermore you don’t limit this statement to supposed Arabic etymologies, but include all etymologies in all dictionaries. If you had said something like, “I’ve noted a consistent problem in a number of English dictionaries when it comes to words with supposed Arabic etymologies...” I doubt anyone would be taking issue. It’s not the specifics of your arguments, but the broad-brush conclusions that are objectionable.

Dave Wilton - 22 November 2010 12:51 PM

The OED lists 597 entries with “Arabic” in the etymology. (Not all of these are from Arabic, some are cognates, etc., but there are clearly more than Wikipedia lists. That would drive your numbers down considerably.

In principle, statisically speaking, if I used a larger sample from a dataset of the same nature it wouldn’t change my numbers. The sample of 150 words is large enough. You’re saying that if I enlarged the sample it wouldn’t enlarge the number of raspberries found. I can’t agree, knowing as I do that I arrived at this sample in an arbitrary way. But I hasten to add that the Wikipedia list excludes words that are tied to obvious associations with Islam or the geography of the Arabic-speaking countries. Like it doesn’t include fatwa, imam, jihad, etc. The Wikipedia page includes falafel and other Arabic cuisine words, but I do not include them in my list of 150 or so words. Obviously, if those obviously Arabic words were to be included, and if that’s what you mean to say, then I agree with you. But I only want to consider words where the origin is not so trivial, and I hope you can agree with me about that.

Dave Wilton - 22 November 2010 12:51 PM

The second is that you are not saying that a select list of dictionaries that you have consulted are at fault. You say that all English dictionaries are unreliable, and furthermore you don’t limit this statement to supposed Arabic etymologies, but include all etymologies in all dictionaries.

The seven dictionaries I named (plus Ayto’s 2005 book makes eight) constitute just about all of the English dictionaries in print today. (Not counting the full OED which is not in the same league as the others.) I’ve said that all these English dictionaries that I could search online are at fault in a significant number of words involving Arabic origin stories. I haven’t generalized beyond that, I haven’t looked into the question beyond that, and I don’t intend to spend time on it beyond that. But I’ll now venture to say: in the course of this exercise with Arabic words, I’ve picked up a feeling for standards and methods at the dictionaries, and I tend to expect that similar standards and methods would be in play in sample sets of words with no connection with Arabic-origin stories.

I tend to expect that similar standards and methods would be in play in sample sets of words with no connection with Arabic-origin stories.

No, not at all. It could be that the dictionaries are relying on the research of one or more Arabic language experts that is faulty, especially if, as is often the case, the work was done decades ago. Arabic-to-English etymology is a pretty specialized field and there is no reason to believe that the Germanic etymologies or the Anglo-Norman etymologies, for example, suffer similar problems.

And, this is very important, we don’t know that these etymologies are actually in error. We don’t know what evidence the dictionaries used to create those etymologies. Only one side has been heard from. And from the tools you used, basic searches of free web resources, it is clear that that there are lots of data sources that we haven’t considered in this thread. There are volumes of books that are only found in research libraries, limited access databases, thousands of journal articles, to be searched for evidence. All we know is that there is a reasonable basis for questioning these specific etymologies as printed in the cited dictionaries.