The Origin of the Later Egyptian Negative Particle iwn3 ~ in ~ AN as a Minimizer “Piece”: A Jespersen’s Cycle Etymology Also Relevant to Scholars of Arabic and Berber

The following is adapted from a fuller article draft that began peer review in Fall 2023, and is available from the author by request.

Rendered in group (syllabic) writing as the sequence transliterated iwn3, in Demotic script as the sequence transliterated in, and in Coptic script as ⲀⲚ an, the Later Egyptian discontinuous negative particle has proven to be surprisingly, frustratingly opaque in its origin.

For over a century (Gardiner 1904), it has been insightfully and properly compared to structures like the French ne… pas, especially through the Coptic-script sequence (Ⲛ︦)…ⲀⲚ (n) … an: as with the French, this negative particle has a later origin than the preceding primary negator, and eventually it even begins appearing as the sole negator.

By implication, especially in light of much subsequent cross-linguistic research into what is now known as “Jespersen’s Cycle” (Willis, Lucas, and Breitbarth 2013: 6–7; Breitbarth, Lucas, and Willis 2020: 10, 36–44), it thus seems very likely that this Egyptian particle had an origin resembling that of the French pas (Alkire and Rosen 2010: 304; Hansen 2013: 53–63). There, the French pas “step” started as an optional direct object to reinforce negation in certain situations. “Help you clean the kitchen? No, I’m not moving a step!” Then, such reinforcement became more-regularly and more-widely used, and it became grammaticalized as a more-general negator.

However, when the Later Egyptian negative particle first appears as the sequence transliterated iwn3, it does so in an already-grammaticalized form, and it is orthographically represented with the group (syllabic) writing that cleaves it from a representation tightly connected to whatever word originated it. Furthermore, when tackling this etymological mystery, the majority of previous proposals nevertheless do *not* follow the fairly strong implication that this Egyptian negative particle appears to be yet another occurrence of the same cross-linguistic pattern, and instead they go in a different direction (Groll 1970: 155; Davis 1973: 192; Meltzer 1990; Winand 1997). Even the other previous proposals that think more closely with such a comparison face problems with a lack of a clearly parallel originating word (Volten 1959; Vycichl 1983: 11). Consequently, it has not only been recently stated that there is “no consensus as to the correct etymology” in regards to “a possible earlier history as a less grammaticalized form” (Lucas 2013: 407n9), but it has also been pessimistically lamented that “the prehistory of iwn3 cannot be traced” (Breitbarth, Lucas, and Willis 2020: 141).

Sadly, this situation has continued despite this Egyptian etymology’s disproportionately large interest to scholars of other Afroasiatic languages, due to ongoing discussion about this particle’s possible influence on some Arabic dialects and eventually some Berber languages (e.g. the positions and debate seen across Lucas 2007; Lucas and Lash 2010; Lucas 2013: 404–419; Kossmann 2013: 332–334; Souag 2018; Breitbarth, Lucas, and Willis 2020: 136–146; Lafkioui and Brugnatelli 2020). For example, it has recently been proposed that native speakers transferred the Egyptian structure to Arabic by seeking a corresponding negator in their second language and locating one in the generalizing negator “thing,” and it was this fateful linguistic turn-of-events that eventually rippled out through some Berber languages (Breitbarth, Lucas, and Willis 2020: 140–146).

To make another attempt at locating an etymological candidate for this Later Egyptian negative particle of outsize importance, then, it seems methodologically most advantageous to initially bypass the particle’s rather unhelpful representation in hieroglyphic-derived scripts like group (syllabic) writing. Instead, we might take inspiration from the situation of linguistic divergence found in a comparable language like French. There, the descendent of the originating word pas “step” continues to exist in identical graphic form as the negator pas that it birthed. Furthermore, the comparable timespans of attestation should lead us to suspect that an originating word could also have survived in Egyptian through the era of Coptic script. Ungrammaticalized precedents for such a French particle appear in the Latin comedic playwright Plautus (Hansen 2013: 54-56), thus forming approximately 2,200 years through the present day. In comparison, the most chronologically-sensitive study of the Egyptian particle highlights early grammaticalized examples from the 19^th Dynasty (Winand 1997: 224, 231-232), approximately 1,700 years from the first widespread use of Coptic script. Add some number of centuries onto those approximately 1,700 years for the necessary grammaticalization, and the timespans begin looking very similar, indeed. Thus, it is fully plausible that when the Coptic script hit a major orthographic “reset button” for the Egyptian language, two interlinked tokens could suddenly become visible in a way that they were not, under hieroglyphic-derived script regimes.

Thus, to follow up on this very plausible but yet-uninvestigated possibility, the task presents itself: is there another Coptic-script word that is not only written identically to the negative particle ⲀⲚ an, but that also bears the appropriate sense predicted by cross-linguistic comparison?

And, the answer does indeed seem to be “Yes!”

For, believe it or not, there exists an archaic and slightly grammatically-misdescribed word ⲀⲚ  an that could have plausibly functioned as a stereotypical minimizer within lexical reinforcement.

Specifically, this other ⲀⲚ an is what is currently being described as a “prefix” in collective numerals, such as ⲀⲚⲦⲀⲒⲞⲨ antaiou “(set of) fifty” from ⲦⲀⲒⲞⲨ taiou “fifty” (Crum 1939: 10, 440).

First, and as cannot be emphasized enough, this ⲀⲚ an not only looks alike, but it bears **exactly** the sense that cross-linguistic comparison leads us to expect. As has already been correctly accepted within the two foremost Coptic etymological dictionaries (Spiegelberg 1899: 21; 1913: 124; Černý 1976: 8; Vycichl 1983: 12), this other ⲀⲚ an is linked to the historic noun that is transliterated ‘ and that bore a sense like “piece” (evidently, it was synchronically and diachronically a bit flexible, like the English “piece” and its appearance in words like “ten-piece” when describing sets).

Thus, one can easily imagine grammaticalization from regular, lexically-reinforced utterances wherein something like “I’m not giving!” became the even stronger “I’m not giving a piece!”

Second, and in a very fun quirk of history that demands mention, this other ⲀⲚ an is actually a reanalysis of the noun transliterated ‘ and the following genitive marker that is traditionally transliterated n.

Postulating reanalysis has been a perceptive contribution of this other ⲀⲚ an’s long-standing etymology (Spiegelberg 1899: 21; 1913: 124; Černý 1976: 8; Vycichl 1983: 12), and this recurring position is also eminently understandable cross-linguistically. Reanalysis across morpheme boundaries is known to occur in high-frequency contexts; for example, the English word adder appeared in Old English as næ̅ddre (like the German Natter “adder”), but, due to confusion around the indefinite article, something like a nadder was conceptually reallocated as something like an adder and so the word known now was birthed (Campbell 2020: 99). Thus, given recurrent phrasings like “set of fifty,” it is already conceivable that something like “setta fifty” could emerge. In fact, this likelihood becomes even stronger when we acknowledge that the collapse of a unit and its ensuing genitive marker is already known from at least one other language – namely, within English, the twentieth century innovation of the mainly British word “cuppa” for “cup of tea” (Oxford English Dictionary 2023, s.v. “cuppa”).

Third, a few Coptic-script attestations already contain a rebuttal to any strong skeptic who might doubt the existence of this other ⲀⲚ an as a truly reanalyzed, self-standing word, and who instead would leadenly parse attestations like ⲀⲚⲦⲀⲒⲞⲨ antaiou “setta fifty” in a contrarian fashion as something like Ⲁ - Ⲛ - ⲦⲀⲒⲞⲨ a - n - taiou. Here, the key issue is whether we have attestation of morphological processes that show the integrity of ⲀⲚ an as a unit, much like how the somewhat parallel British word “cuppa” can be pluralized and appear in phrases like “community cuppas” (versus the ungrammatical *“community cup ofs”) (Murray and Pidd 2020). Thus, highly relevant are what has been termed “reiterations” in a current Coptic grammar, since these reiterations like ϨⲞⲞⲨ ϨⲞⲞⲨ hoou hoou “one day after another” and ⲘⲀ ⲘⲀ ma ma “one place after another” presume a single core word that is then doubled up (Layton 2011: 175). And, in fact, we have at least three examples of reiteration of this other ⲀⲚ an in biblical translations. One example appears in the translation of the Festival of “Weeks” with ⲚⲒⲀ́ⲚⲀⲚⲌ︦ nianan7 “the sets of 7” (Exodus 34:22 in Lagarde 1867: 214, and also cited in Crum 1939: 10 and Vycichl 1983: 12). Then, there appears the twin pairing of ⲚⲈϤⲀⲚⲀⲚϢⲞ ne-f-an-an-šo “his thousands” and ⲚⲈϤⲀⲚⲀⲚⲐⲂⲀ ne-f-an-an-thba “his ten thousands” in the famous, jealousy-stoking proclamation of women of Israel that “Saul has killed his thousands, and David his ten thousands” (NRSV; 1 Kings 18:7 in Lagarde 1879-1880: 70, and also cited in Crum 1939: 10, Černý 1976: 8, and Vycichl 1983: 12). In the reiteration of ⲀⲚ an found in all of these places, there can be perceived a structure that historically resembles something like “setta setta seven,” setta setta thousand,” and “setta setta ten thousand” – a structure, very importantly, that shows the existence of an integral word ⲀⲚ an that resonates with the sense necessary to birth the Later Egyptian negative particle.

As one minor implication of the foregoing analysis, one should probably adjust the description of this other AN as a “prefix,” and instead see it as an archaic noun in a historic compound noun construction of extremely limited distribution (Loprieno 1995: 56–57). Tellingly, other “prefix”-like morphemes like the agent-former ⲢⲈϤ- ref do not seem susceptible to reiteration in sequences like *ⲢⲈϤⲢⲈϤ refref (Layton 2011: 52–53, 81, 175, 214–215, 230–231, 359; Coptic SCRIPTorium 2023), further bolstering this conceptual and terminological adjustment.

As another minor implication of the foregoing analysis, one should also note several issues dealing with orthographic representation of sounds. First and foremost, connecting the Later Egyptian negative particle to this reanalyzed noun may help explain the closing -n3 portion of the group (syllabic) writing sequence transliterated iwn3. Rarer Coptic-script tokens like ⲚⲀⲘⲈ name “truly” (< n m3‘t “for truth”) and ϪⲈⲚⲈⲠⲰⲢ jenepōr “roof” (< d̠3d̠3 n pr “head of the house”) (Černý 1976: 78, 316) suggest that the genitive marker once appeared as /nV/ between nouns. Thus, it would be plausible for this closing vowel of the genitive marker to have carried over into the reanalyzed noun and the negative particle that it birthed. In turn, it thus would also have been very plausible for the final /nV/ part of a phonetic shape like /VnV/ to have possibly been represented by the group (syllabic) writing sequence transliterated n3 (cf. tokens in Kilani 2019: 115–116). Here, of course, one would then be forced to posit some sort of final vowel loss that took place after this negative particle first appeared in group (syllabic) writing. The current etymology also raises a different, more diffuse question about the relation of the word transliterated ‘ to what became transliterated as iw, particularly in relation to previous claims about how a sound corresponding to ‘ persisted to some degree through the era of Demotic script (Carsten Peust, p.c..; Peust 1999: 99–100, 102–103; Allen 2020: 29–30).

Lastly, there is the matter of earlier attestations stemming from this Coptic-derived prediction – that is, appearances of congruent, less-grammaticalized reinforcement with the word “piece” that is transliterated ‘, such as in phrasings like “I’m not giving a piece!” To think broadly about this issue, and as is resonant with previous suspicions that the Later Egyptian negative particle emerged from the Volkssprache (Meltzer 1990: 75), such attestations would most likely surface in marginal, non-monumental texts that contain or represent informal speech – in other words, the very texts that are very unlikely to survive in great numbers, if at all.

In fact, after hypothesis formation, a preliminary search of more than 600 token occurrences in a major multitext database (Thesaurus Linguae Aegyptiae 2023) turned up only two possible relevant usages, then luck perhaps a third, in the pages of a standard Egyptian grammatical resource (Gardiner 1953: 342).

One of the earliest known usages seems to be in the Lahun papyri, in a copy of a 12^th Dynasty letter where the word “piece” appears in a prepositional phrase meaning something like “to a piece” or “completely” (…mi nt.t nb.t r ‘ n nb… “…as everything is completely for the lord” in Papyrus Cairo JE 71581, most recently edited in Luft 2006: 113-117). A subject worthy of longer treatment in and of itself, this usage probably resulted in the later Coptic-script particle ⲢⲰ rō (Crum 1939: 290-291; Shisha-Halevy 1990: 120–121; Depuydt 2001: 119–123). Although certainly not the same usage that led to the formation of the Later Egyptian negative particle, this example nevertheless suggests that that important word “piece” was surfacing in the appropriate time period in at least two different polarity items (i.e., language-specific expressions that reference scales of possibility and can express maximal extent in positive or negative contexts or both, per Israel 2011: 7, 38-41, 43-47).

Much more germane is a textual variant from the Tale of the Eloquent Peasant that has been dated to the 13^th Dynasty (18.1 and 18.4-5 in pBerlin P 10499, also known as P Ramesseum A; for probable dating, see Parkinson 2012: 1-3). In a bit of dialogue from a regrettably fragmentary manuscript of this famous piece of Egyptian literature, the idiom “give provision” (ir ‘nh̬) found elsewhere is instead rendered as “give a bit of provision” (ir ‘ n ‘nh̬) (Vogelsang and Gardiner 1908: Table 4; Parkinson 1991: 20, 21; 2012: 90-91). In other words, not only does this manuscript variant depict spoken language, but it also almost certainly corresponds to an actual historic phrasing option available to native speakers of Egyptian – specifically, the ability to act deignfully and state “give a bit of provision,” vs. simply saying “give provision.” Very pointedly, this phrasing alternation shows that native speakers were deploying the positive-context counterpart of the presumably more-common negative context usage that went on to birth the negative particle (i.e., “I’ll give you a piece,” vs. “I’m not giving you a piece!”).

Finally, an 18^th Dynasty of Thutmose II (Sethe 1906: 138; Gardiner 1953: 342) may preserve “piece” in the crucial, ambiguous negative context, where it can possibly be interpreted as something like “bit” or as the negative particle (n h̬sf ‘ n ipwty-f… “one does not impede his messenger a bit…”).

If all of these three examples have been interpreted correctly – and other respectable interpretations are possible! – they thus collectively show multiple polarity items using “piece” occurring in the 12^th and 13^th Dynasties, and then incipient grammaticalization during the 18^th Dynasty, immediately prior to the already-grammaticalized 19^th Dynasty examples that appear in origin-disguising group (syllabic) writing. Strikingly, these examples invoke approximately 2,200 years from the 12^th Dynasty through the first widespread use of Coptic script that reveals the divergence situation with ⲀⲚ an, a timeframe roughly paralleling that from the precedents of pas in Plautus through modern French. Moreover, the Tale of the Eloquent Peasant neatly parallels Plautus as a source-type; in both genres, literary depictions of spoken speech preserve the crucial early examples.

Indeed, even if all of these examples are not accepted, the example from the Tale of the Eloquent Peasant seems like fairly decisive proof for the origins of the Later Egyptian negative particle in a minimizer “piece.” Admittedly, it is a quirk of textual survival that it bears witness to a positive context, rather than the negative context usage that was probably more widespread. Yet, if one takes a roughly contemporaneous negative construction from another version of the same text (n irr-k st “You will not do it” of B1 152, as in Parkinson 1991: 24 and Parkinson 2012: 129), replacement of the direct object with the attested phrasing option produces an ambiguous grammatical structure easily permitting reinterpretation of “piece” as a negative particle (i.e., the artificial example *n irr-k ‘ n ‘nh̬ could be interpreted as “You will not give a-bit-of provision,” but it could also be interpreted as “You will not give provision,” with “a-bit-of” understood as discontinuous negation like that of Later Egyptian; cf. the thought experiment of Breitbarth, Lucas, and Willis 2020: 142–143).

In any case, scholars who work with plausible corpuses from at least half a millennium before the 19^th Dynasty should be on the lookout for minimizing usage of this word transliterated ‘ and meaning “piece,” especially in negative contexts, for it is fully expected from the evidence laid out here. If found, such evidence would help further establish the present Jespersen’s Cycle etymology as the most elegant solution to this enduring Egyptian mystery that also concerns many other scholars working in Afroasiatic.

. . .

David Mihalyfy is an independent scholar with a B.A. in Linguistics (Harvard, ’02) and a Ph.D. in the History of Christianity (UChicago ’17).

In a not uncommon scholarly trajectory, he first became interested in Afroasiatic historical linguistics because he happened to study one language (Coptic, for Christian texts), and soon realized the relative underdevelopment of diachronic knowledge in comparison to Indo-European languages.

He has presented at several Egyptological Symposiums of the American Research Center in Egypt’s Missouri Chapter, as well as remotely at the Egyptological Conference in Copenhagen. Several short articles from a larger project on the survival and transformation of Afroasiatic N-stems in Egyptian were recently published in the Journal of the Society for the Study of Egyptian Antiquities.

. . .

WORKS CITED.

Ti Alkire and Carol Rosen, Romance Languages: A Historical Introduction (Cambridge: Cambridge University Press, 2010).

James P. Allen, Ancient Egyptian Phonology (Cambridge: Cambridge University Press, 2020).

Anne Breitbarth, Christopher Lucas, and David Willis, The History of Negation in the Languages of Europe and the Mediterranean, Volume II: Patterns and Processes (Oxford: Oxford University Press, 2020).

Lyle Campbell, Historical Linguistics: An Introduction, 4^th ed. (Cambridge, MA: The MIT Press, 2020).

Jaroslav Černý (ed.), Coptic Etymological Dictionary (Cambridge: Cambridge University Press, 1976).

Coptic SCRIPTorium (2013-2023), edited by Caroline T. Schroeder, Amir Zeldes, et al., ANNIS search of every text corpus for examples of “ⲣⲉϥ” .1,3 “ⲣⲉϥ”, https://copticscriptorium.org (accessed 25 October 2023).

W.E. Crum, A Coptic Dictionary (Oxford: The Clarendon Press, 1939; Eugene, OR: Wipf and Stock, 2005 reprint).

Viriginia Lee Davis, Syntax of the Negative Particles bw and bn in Late Egyptian (Munich: Deutscher Kunstverlag, 1973).

Leo Depuydt, “The Meaning of the Coptic Particle ⲣⲱ and Related Constructions in Semitic and Other Languages,” Journal of Coptic Studies 3 (2001): 113–128.

Adolf Erman and Hermann Grapow, Wōrterbuch der aegyptischen Sprache, 2^nd vol. (Leipzig: J.C. Hinrichs’sche Buchhandlung, 1928).

Alan Gardiner, “The word [iwn3],” Zeitschrift für Ägyptische Sprache und Altertumskunde 41 (1904): 130–135.

Alan Gardiner, Egyptian Grammar: Being an Introduction to the Study of Hieroglyphs, 3^rd ed. (Oxford: Griffith Institute, 1953).

Sarah Israelit Groll, The Negative Verbal System of Late Egyptian (Oxford: Griffith Institute, 1970).

Maj-Britt Mosegaard Hansen, “Negation in the history of French,” in David Willis, Christopher Lucas, and Anne Breitbarth (eds.), The History of Negation in the Languages of Europe and the Mediterranean, Volume I: Case Studies (Oxford: Oxford University Press, 2013), 51-76.

Michael Israel, The Grammar of Polarity: Pragmatics, Sensitivity, and the Logic of Scales (Cambridge: Cambridge University Press, 2011).

Marwan Kilani, Vocalisation in Group Writing: A New Proposal (Hamburg: Widmaier Verlag, 2019).

Maarten Kossmann, The Arabic Influence on Northern Berber (Leiden and Boston: Brill, 2013).

Mena B. Lafkioui and Vermondo Brugnatelli, “Negation in Berber: Variation, evolution, and typology,” Linguistics 58(4) (2020): 967-1008.                                                                             

Paul de Lagarde, Der Pentateuch Koptisch (Leipzig: B.G. Teubner, 1867).

 ------, Orientalia, erschienen in: Abhandlungen der Kōniglichen Gesellschaft der Wissenschaften zu Gōttingen Band XXIV und XXVI (1879-1880; Osnabrück: Otto Zeller Verlag, 1973 reprint).

Bentley Layton, A Coptic Grammar, 3^rd ed. (Wiesbaden: Harrassowitz Verlag, 2020).

Antonio Loprieno, Ancient Egyptian: A Linguistic Introduction (Cambridge: Cambridge University Press, 1995).

Christopher Lucas, “Jespersen’s Cycle in Arabic and Berber,” Transactions of the Philological Society 105(3) (2007): 398–431.

------, “Negation in the history of Arabic and Afro-Asiatic,” in David Willis, Christopher Lucas, and Anne Breitbarth (eds.), The History of Negation in the Languages of Europe and the Mediterranean, Volume I: Case Studies (Oxford: Oxford University Press, 2013), 399-452.

Christopher Lucas and Elliott Lash, “Contact as catalyst: The case for Coptic influence in the development of Arabic negation,” Journal of Linguistics 46 (2010): 379–413.

Ulrich Luft, Urkunden zur Chronologie der spāten 12. Dynastie: Briefe aus Illahun (Vienna: Verlag der Österreichischen Akademie der Wissenschaften, 2006).

Edmund S. Meltzer, “The ‘Prehistory’ of Late Egyptian i(w)n(3): A Hypothesis,” Göttinger Miszellen 114 (1990): 71–79.

Oxford English Dictionary, “cuppa, n., Etymology,” July 2023 version, https://doi.org/10.1093/OED/7118830897 (accessed 16 October 2023).R.B. Parkinson, The Tale of the Eloquent Peasant: A Reader’s Commentary (Hamburg: Widmaier Verlag, 2012).

R.B. Parkinson (ed.), The Tale of the Eloquent Peasant (Oxford: Griffith Institute, 1991).

Jessica Murray and Helen Pidd, “Coronavirus: looking for good news - community cuppas and the Stockport Spiderman,” The Guardian, 31 March 2020, https://www.theguardian.com/uk-news/2020/mar/31/coronavirus-looking-for-good-news-and-hope-in-a-time-of-crisis (accessed 16 October 2023).

Carsten Peust, Egyptian Phonology: An Introduction to the Phonology of a Dead Language (Göttingen: Peust & Gutschmidt Verlag, 1999).

Kurt Sethe (ed.), Urkunden der 18. Dynastie, Vol. 2: Historisch-biographische Urkunden aus der Zeit der Kōnige Thutmosis’ I und II (Leipzig: J.C. Hinrichs’sche Buchhandlung, 1906).

Ariel Shisha-Halevy, “The ‘Tautological Infinitive’ in Coptic: A Structural Examination,” Journal of Coptic Studies 1 (1990): 99–127.

Lameen Souag, “Arabic-Berber-Songhay contact and the grammaticalisation of ‘thing,’” in Stefano Manfredi and Mauro Tosco (eds.), Arabic in Contact (Amsterdam: John Benjamins Publishing Company, 2018), 53–71.

Wilhelm Spiegelberg, “Koptische Miscellen,” Recueil de travaux relatifs à la philologie et à l’archéologie égyptiennes et assyriennes 21 (1899): 21–22.

------, “Die allgemeine Orts- und Zeitbestimmung ‘ im Kopstischen,” Zeitschrift für Ägyptische Sprache und Altertumskunde 41 (1913): 122–126.

Thesaurus Linguae Aegyptiae, Corpus issue 18 / Web app version 2.1.2 (24 November 2023), edited by Tonio Sebastian Richter and Daniel A. Werning on behalf of the Berlin-Brandenburgische Akademie der Wissenschaften, and edited by Hans-Werner Fischer-Elfert and Peter Dils on behalf of the Sächsische Akademie der Wissenschaften zu Leipzig, searches for hieroglyphic/hieratic occurrences of lemma 34360 and lemma 34340, https://thesaurus-linguae-aegyptiae.de (accessed 13 February 2024).

F. Vogelsang and Alan H. Gardiner (eds.), Literarische Texte des Mittleren Reiches, Vol. 4: Die Klagen des Bauern (Leipzig: J.C. Hinrichs’sche Buchhandlung, 1908).

Aksel Volten, “L’étymologie de deux mots coptes,” Bulletin de l’Institut Français d’Archéologie Orientale 58 (1959): 21–28.

Werner Vycichl, Dictionnaire Étymologique de la Langue Copte (Leuven: Peeters, 1983).

David Willis, Christopher Lucas, and Anne Breitbarth, “Comparing diachronies of negation,” in David Willis, Christopher Lucas, Anne Breitbarth (eds.), The History of Negation in the Languages of Europe and the Mediterranean, Volume I: Case Studies (Oxford: Oxford University Press, 2013), 1-50.

Jean Winand, “La négation bn… iwn3 en néo-égyptien”, Lingua Aegyptia 5 (1997): 223–236.