Paximadia.

I was flipping through Alan Davidson’s Penguin Companion to Food (see this 2005 post) when my eye fell on the entry Paximadia, “an exceptionally interesting Greek item in the frontier area between breads and biscuits.” It ends:

Paximadia do not belong only to Greece. Kremezi’s essay and Dalby (1996) between them illuminate their wider distribution and the likely derivation of their interesting name.

Which of course sent me on a quest. Wikipedia says:

The name paximathia comes from the Greek term paximadion (Greek: παξιμάδιον), which is derived from Paxamus, a 1st-century Greek author who wrote, among many things, a comprehensive cookbook. The word first appears in a recipe for laxative biscuits composed by the Greek physician Galen.

Wiktionary says “Ultimately from Ancient Greek παξαμᾶς (paxamâs, ‘biscuit’) and says the latter is “From Πάξαμος (Páxamos), the name of a baker.” I’m not saying that’s wrong, but it has a whiff of folk etymology to me, and I’m wondering how convincing it is to others. (Also, if anyone has tried them, are they perceptibly different from your basic rusk?)

Comments

  1. In modern Turkey, the city of Mardin is well known for preserving the tradition of peksimet (from Modern Greek παξιμάδι, ancient παξαμάδιον). There is an article here (run through your preferred translator) with a link to this YouTube video of a typical shop in Mardin. There are several wood-fired bakeries with stone ovens in every neighborhood in Mardin, and on my way to and from work I pass tables laden with big bags of peksimet put out on the street for sale. Peksimet are made from a sort of whole-wheat flour and are not sweetened. I prefer them softened in soup or sauce.

    The derivation of παξαμᾶς from Πάξαμος does not trouble me particularly. Offhand, as a parallel, I can think of Catalan maçana, Portuguese maçã, Spanish manzana from *(māla) mattiāna, probably after Gaius Matius. Cf. Suetonius (Life of Domitian, 21):

    Quotiens otium esset, alea se oblectabat, etiam profestis diebus matutinisque horis, ac lavabat de die, prandebatque ad satietatem, ut non temere super cenam praeter Matianum malum et modicam in ampulla potiunculam sumeret. Convivabatur frequenter ac large, sed paene raptim; certe non ultra solis occasum nec ut postea comissaretur. Nam ad horam somni nihil aliud quam solus secreto deambulabat.

    Whenever he had leisure he amused himself with playing at dice, even on working days and in the morning hours. He went to the bath before the end of the forenoon and lunched to the point of satiety, so that at dinner he rarely took anything except a Matian apple and a moderate amount of wine from a jug. He gave numerous and generous banquets, but usually ended them early; in no case did he protract them beyond sunset, or follow them by a drinking bout. In fact, he did nothing until the hour for retiring except walk alone in a retired place.

    The personal name Πάξαμος is odd. I would love to know more about it. The possibility of an Egyptian origin has been floated, but what would the Egyptian etymon then be?

  2. I’ve made paximadia from recipes I found online, with a mix of barley flour and whole wheat. Barley has distinctive kind of malty flavour and while I’m not sure what a “basic rusk” is, I assume this would be different.. Texturally they do make a great base for a well-dressed salad, the contrast is nice and so are the wet crumbs that soak up the moisture from the salad. But watch out if you have weak teeth.

  3. The derivation of παξαμᾶς from Πάξαμος does not trouble me particularly.

    Well, if it doesn’t bother you, I won’t let it bother me! And of course I appreciate the additional research.

  4. How did the α of παξαμᾶς become the first ι of παξιμάδιον?

    How does a malus matianus differ from any ol’ malus?

    παξαμᾶς appears as פַּכְסָם paḵsām in Midrash Bamidbar Rabbah 7, 4: “Let the days be remembered when we were in Egypt, while we were cooking pot after pot of meat, and sitting and eating paḵsāmîn, and crumbling them into the meat juice.” In Israel the word was adopted for rusk, Swedish rye crackers, and the like, but I don’t know if it’s in much common usage.

  5. in egypt we too say bu’sumat . th q sound became a glottal stop i guess

  6. How did the α of παξαμᾶς become the first ι of παξιμάδιον?

    That was one of the things that bothered me.

  7. David Marjanović says

    th q sound became a glottal stop i guess

    That’s normal in Egypt.

  8. Johann Knobloch, Romance Philology 37 (1983), p. 80 (in a review of R. and H. Kahane, Abendland und Byzanz: Sprache, part of the Reallexikon der Byzantistik), suggests an entirely different etymology, deriving it from *ἀποξαμάτη, a reference to the double baking process, in which the baked dough is pulled apart, rewetted, and baked again:

    Uber παξαμᾶς ‘Biskuit’ weiß Liddell-Scott, daß es einen Bäcker namens Paxamos gegeben habe. Aber das Wort spiegelt ein bekanntes Backverfahren: gebackener Teig wird zerrissen und mit Flüssigkeit versetzt ein zweites Mal gebacken; παξαμάτης ‘Zwieback’, ital. pasimata ‘Kuchenbrot’ und ngr. παξιμάδι sind also als *ἀποξαμάτη zu deuten und nahe Verwandte von ἀναξασμός ‘das Zerfetzen’, zum Verbum ἀποξαμάω ‘reibe ab’ zu stellen.

  9. Aha! I knew the Paxamos story smelled off (like the Crapper pseudo-etymology)! I like that Knobloch fellow.

  10. He must have been the guy who brought garlic to Germany.

  11. In Bosnian this is “peksimet” although the word has taken on a broader meaning to include things like fritters. The native slavic word is “dvopek” (i.e. twice-baked)

  12. Is dvopek taken to be a calque of biscoctus maybe by way of Zwieback?

  13. Thank you MEL for that reference! I appreciate everything I can learn about the history of Turkish peksimet, Mardini colloquial Arabic bakısma, basic words of the city of Mardin.

    I am travelling and away from my library and I would appreciate it if someone could point me to some references to this verb ἀποξαμάω. I couldn’t find it in any online resources readily accessible to me now, including the LSJ and Lexikon zur byzantinischen Gräzität online or the Λεξικό της Μεσαιωνικής Ελληνικής Δημώδους Γραμματείας 1100-1669. (I would have expected it to be here in the ΛΜΕΔΓ.) I am confused by the reference to ἀναξασμός ‘a rip, tear’, which is derived from ἀναξαίνω ‘wieder aufkratzen, wieder erneuern, refricare; im pass., von Wunden, wieder aufbrechen’, beside this verb (there is an ἀποξαίνω known in the passive as ‘be torn’).

    The Byzantine forms ἀπαξαμᾶς, ἀπαξαμάτος, ἀπαξιμάδιον are here in the LBG. Already in 1892, Georgios Hatzidakis in his Einleitung in die neugriechische Grammatik seems to attribute the initial ἀ- in these forms not to persistence of the original form derived from a verb with ἀπ(ο)- like Knobloch, but rather to an opposing process, folk-etymological reanalyis of the forms like παξαµάτια as containing an aphetic version of the prefix ἀπ(ο)- (see p. 375 here), with attempts at restoration of ἀπ(ο)- variously.

    At its entry for παξιμάδιν, the ΛΜΕΔΓ (vol. 14, 1997) takes the koine form παξαμᾶς as original (here, top of p. 308) and seems to follow Hatzidakis in considering the ἀ- to be prothetic.

  14. The farthest back that I have been able to trace the etymology explicitly taking παξαμᾶς from Πάξαμος is to a passage (apparently consisting of some extracts attributed to Julius Pollux) in a 14th century manuscript now in the Bibliothèque nationale de France, Paris 1630. An image of the relevant page (fol. 95v) is on the left hand side here. The passage begins in line 2. The passage is edited on page 267 (here) of Emmanuel Miller (1867) ‘Extraits de l’Onomasticon de J. Pollux’ Revue archéologique, vol. 27:

    Ὅτι ὁ Πάζαμος εἷς ἦν τῶν σοφιστῶν τῆς ὀψοποιητικῆς πραγµατείας, καὶ
    ὥς ἀπ’ ἐχείνου οἶμαι ἐκλήθησαν τὰ παξαµάτια.

    Paxamos was one of the masters of the business of cookery, and so it is after him, I think, that παξαμάτια were named.

    For more on this manuscript, see the introduction of Miller’s article on p. 260. All I have time for right now—but again, I appreciate everything I can learn about the history of this word. (A previous comment of mine has fallen into moderation, I hope it reappears.)

  15. Correct to Ὅτι ὁ Πάξαμος… Apologies for not catching this OCR error.

  16. Doesn’t the asterisk mean that Knobloch hypothesizes this verb, which is unattested?

  17. In the text as quoted in this thread, only the noun ἀποξαμάτη is marked with an asterisk, not the verb:

    ngr. παξιμάδι sind also als *ἀποξαμάτη zu deuten und nahe Verwandte von ἀναξασμός ‘das Zerfetzen’, zum Verbum ἀποξαμάω ‘reibe ab’ zu stellen.

    The only hit with a Google search for
    ἀποξαμάω is actually this thread.

  18. There are many things that puzzle me about Knobloch’s note.

    I was wondering whether it was an ἀπο- + ἐξ- + ἀμάω ‘mow, reap, cut’ (for a virtual *ἀπεξαμάω) that was meant.

    Also, the gloss that appears in Knobloch’s note, reibe ab (in a culinary context, ‘grate’?), would not be appropriate for such a formation. I wonder, is reibe ab (printed so, I checked on JSTOR) actually a misprint for reiße ab ‘tear off’? (Cf. zerrissen earlier in the note.)

    I would also like to see a fuller historical recipe showing this exact process of preparation (“pulled apart, rewetted, and baked again”)—I wonder, how does this not produce something with a jagged, crumbly surface that would not be ideal for storage, or for transport in horse and camel packs? The biscuits I know are all sliced cleanly and then simply baked again (rusks, biscotti, Turkish peksimet) or otherwise shaped to be regular (hardtack), but of course preparations in earlier times may have been different.

  19. Here (from this article) is an image of some peksimet as it is traditionally made, it seems, in Bodrum, Turkey. These might be considered ‘torn off’ (that is, it looks like the dough was baked into loaves, the loaves were torn up, and then the torn pieces were dried in the oven). I suppose such pieces can be more densely packed in storage. Maybe it improves the flavor, or there is greater surface area available when it is moistened with water or soup.

  20. I was wondering whether it was an ἀπο- + ἐξ- + ἀμάω ‘mow, reap, cut’ (for a virtual *ἀπεξαμάω) that was meant
    That would be my guess as well. The question is whether this verb is actually attested or whether Knobloch assumes it (but then it should have been asterisked).

  21. I do not know whether they add much, but here are du Cange Latin Greek and Koukoules.

  22. Where does biksemad come from? The ingredients are wrong, of course. ODS doesn’t seem to explain where the sailors got it.

  23. Trond Engen says

    I was pondering a joke on that.

    Norw. bikse is a word of ON pedigree for “rich and powerful person”, Da. -mad is just “food” (but you knew that), so “food for the mighty”, applied to a meal made from leftovers in the kitchen (or rather galley). I guess there could be an element of folk-etymology, if so presumably from a Dutch or English original. The only word that comes to mind is Ger. bisschen “tiny” < “little bit”, but that doesn’t work in LG/Dutch.

  24. I was wondering about Danish sailors with their biksemad encountering Turkish sailors with their beksemad (peksimet).

  25. This is Knobloch’s review article:
    https://www.jstor.org/stable/44942555

    He uses asterisked forms quite a bit.

  26. Speisen mit dem Kaiser von Byzanz: Antike Gaumenfreuden aus Küche und Keller — verkostet von Liutprand von Cremona” has this:

    Als griechisches Synonym für paximacium läßt sich das fast gleichlautende paximadi(on) bestimmen. Die Etymologie ist umstritten, am frühesten läßt sich das Wort in den medizinischen Schriften Galens nachweisen. Ob es mit dem Namen des böotischen Geoponikers Paxamos in Verbindung zu bringen ist, der im 1. Jh. n. Chr. lebte und dessen Schriften in der bereits mehrfach herangezogenen Geoponika des 10. Jhs. n. Chr. häufig zitiert werden, oder ob der Ursprung dieser Bezeichnung für Zwieback im lateinischen Wortschatz zu suchen ist, bleibt als philologisches Problem ebenso ungelöst wie eine mutmaßlich ägyptisch-koptische Provenienz.

    All but the last clause has been covered before: Galen, obscure etymology, maybe named for a Paxamos. But is a Coptic origin something suggested by someone someplace or a philologist’s joke? I believe that AW, like BAR, doesn’t forbid footnotes, but neither does it require them, in favor of glossy photos.

  27. Stu Clayton says

    Geoponika

    A grandfather of terraforming and ecosynthesis.

  28. Trond Engen says

    I don’t think there’s any reasonable route for a specifically Turkish form of the name to end up in Danish without any intermediates. I see two unreasonable routes:

    1. Danish sailors returning from captivity in the Barbary. That’s Lameen’s neck of the woods
    .
    2. The 17th C. Norwegian cadet turned Dutch navy officer turned Venetian admiral fighting the Turks in the Aegean, Cort Adeler. After retiring from Venetian service he was called home from Holland by the Danish king to command and modernize the Danish fleet.

    Neither route explains the change in meaning.

  29. ebenso ungelöst wie eine mutmaßlich ägyptisch-koptische Provenienz

    Such an origin is proposed by Ignazio Rossi (1808) Etymologiae Aegyptiacae, starting at the end of page 155 here. The Egyptian antecedent would be the equivalent of a Coptic ⲡ-, masc. sg. definite article + ϣⲏⲙ ‘small’ + ⲟⲉⲓⲕ ‘bread, loaf’ in his account. I do not have time now to explore the morphological and Coptic dialectological details of his proposal—maybe someone can follow up.

  30. For lagniappe… A remarkable proposal for an Egyptian etymology of another Latin word for a kind of baked good, torta, turta, again from the Kahanes whose work Knobloch was reviewing, is here (Henry and Renée Kahane, “The Role of the Papyri in Etymological Reconstruction.” Illinois Classical Studies 3 (1978): 207–20).

  31. Another remarkable proposal from the same paper, likewise well-argued, is that peridot is a doublet of pederast.

  32. Egyptian antecedent

    Without presuming to judge this reconstruction, taking it in the other direction:

    ϣⲏⲙ < ḫm Er 359.3 Wb 3, 281.13. ⲟⲉⲓⲕ < ꜥḳw Er 73,1 Wb 1, 232.16

  33. But note that per Černý the Middle Egyptian ḫmꜣꜣ linked to is only “connected with” the later shorter form.

  34. The edit doesn’t destroy anything, it just looks that way on your side. Refresh and you’ll see that it has come out fine.

  35. One can download the Kahanes’ earlier (1967; also somewhat longer) “Five Romance Etymologies,” containing “The Story of the Torte,” from the rather quirky ELiPhi numérique here. It’s in RLiR 31.

  36. I believe the traditional analysis of the offering list item trtḥ is ‘baker’s bread’. For example, here. It’s possible they are trying to be more careful about their grammar than others were, since the meanings are, of course, very close.

  37. Budge, who wasn’t very reliable even in his own time, but which we all grew up on thanks to Dover Pubs, has reteḥ ‘a kind of sacrificial cake’. Like Hassan above, he points to where we can actually see it in such an offering list from Unas’s Pyramid Texts, Utterance No. 89 (fifth from the right). The English on that site just does tareteh-bread. Allen seems to find some play on words with the Eye of Horus, since he translated as ‘pulled bread’.

  38. David Marjanović says

    reibe ab (in a culinary context, ‘grate’?)

    That is actually just reiben, not abreiben.

    But is a Coptic origin something suggested by someone someplace or a philologist’s joke?

    It’s cited as serious here in any case. Mutmaßlich is somewhere around “putative”; it’s also used by journalists reporting on criminal defendants.

  39. The recent post on hieroglyphics reminded me of this post. I ought to do an outreach talk for the community in Mardin, Turkey, on this topic—no new results, just a summary of the history of peksimet as it as known and the etymological proposals out there, unless something Egyptian pans out.

    That is actually just reiben, not abreiben.

    I was trying to find a meaning for Knobloch’s ἀποξαμάω ‘reibe ab’ that would make sense in relation to rusks, as in the entries in Duden (‘[die Schale von etwas] mit dem Reibeisen entfernen. Beispeile: eine Muskatnuss abreiben; abgeriebene Zitronenschale’) and DWDS (‘die Außenschicht mit dem Reibeisen von einer Frucht reiben’) and the translation offered by Langenscheidt (‘grate’), so grate as in grate a nutmeg; grated lemon zest. But even ‘grate’ doesn’t jibe with Knobloch’s account, since rusks are grated only in their finished state, as in the preparation of a gratin, or to add body to a pudding or a sauce, or a crumbly layer to a crumble—to take some examples from 19th century recipes. The other meanings of abreiben would make even less sense. Hence my proposal above to read reiße ab.

    The derivation of παξαμᾶς from Πάξαμος does not trouble me particularly.

    Looking into an Egyptian origin for παξαμᾶς and παξαμάδιον again, what does trouble me is the formation with -ᾶς.

    This suffix typically forms nouns denoting professions, as well as nouns with a colloquial flavor that denote a person disposed to a certain activity or characteristic. See paragraph e p. 461 in the top of page 32 in Chantraine (top of p. 32) and Schwyzer (who mentioins παξαμᾶς ‘Zwieback’). When meaning ‘rusk’ or the like, παξαμᾶς really stands out here.

    But παξαμᾶς is attested in the sense of ‘baker of paximadia’ in relatively early texts. For example there is the ostracon from Abu Mena 8612 C790b:

    Μηνᾶς παξαμᾶς
    ὀν(ικαὶ) φο̣ρ(αὶ) ιβ μό(νον)

    Menas the biscuit-baker
    12 donkey loads only

    Or something like that. Abu Mena was a pilgrimage site in a wine-producing region. The deposit of ostraca that were found there are mostly receipts given by a local winery given to named individuals in exchange for loads of grapes, carried on donkey or camel back. Apparently the receipts could be redeemed later for the wine when it was ready, and the winery retained the redeemed ostraca in baskets as part of a record-keeping system. All the ostraca include an indication of a date in the Coptic months Mesore or Thoth, the time of the grape harvest in Egypt (roughly August and September). The ostraca date from the second half of the 6th century or the first half of the 7th century. They were apparently deposited in a rubbish tip near the winery soon after the Arab conquest of Egypt in 642 CE, at a time when wineries were reestablished after the peried of disruption and the records of the previous establishment were of no interest to the new wineries. Two other individuals, Ἰσαείας and Ἰωάνης, are described as παξαμᾶς in the ostraca from Abu Mena. Other ostraca mention a Μηνᾶς ἀρτοκόπος ‘Menas the bread-baker’. The same Μηνᾶς as Μηνᾶς παξαμᾶς?

    Another occurrence of παξαμᾶς is in a mosaic (from 573 CE?) found in Sidon described on p. 300ff here. See page 303–304 for the interpretation by as a (kind of) mosaist, building off the derivation of παξαμάδιον from Πάξαμος by Ioannis Psycharis. The entries in the Suda that are referred to there are Πάξαμος and παξαμᾶς (change the encoding to Unicode in the top pull-down menu to display the Greek properly).

  40. David Marjanović says

    Ah. I’m used to grating whole nutmegs, and that’s reiben; agreed on the lemons, though.

Speak Your Mind

*