Home » Books » Mosaic or mosaic? – The Genesis of the Israeli Language

Language Revival and Multiple Causation: The mosaic Genesis of the Israeli Language, Oxford University Press, Forthcoming

Hebrew was spoken after the so-called conquest of Israel (c. thirteenth century BC). Following a gradual decline (even Jesus, ‘King of the Jews’ was a native speaker of Aramaic rather than Hebrew), it ceased to be spoken during the second century AD. The Bar-Kokhba Revolt against the Romans, which took place in Judaea in AD 132-5, marks the symbolic end of the period of spoken Hebrew. According to the Roman historian Dio Cassius (who might have exaggerated), the Romans killed 580 000 Jews, in addition to those who died of hunger, disease and fire, and Bar-Kokhba himself met his death in AD 135 during the fall of Bethar. In the period of repression which followed, the Jewish population in Judaea was largely exterminated through massacres, religious persecution, slavery and forced relocation. For more than 1700 years thereafter, Hebrew was comatose – either a ‘sleeping beauty’ or ‘walking dead’. It served as a liturgical and literary language and occasionally also as a lingua franca for Jews of the Diaspora, but not as a mother tongue. The formation of Israeli (the name I use for so-called ‘Modern Hebrew’) was facilitated at the end of the nineteenth century by Eliezer Ben-Yehuda, school teachers and others to further the Zionist cause. Earlier, during the Haskalah (enlightenment) period of the 1770s-1880s, writers such as Méndele Mokhér Sfarím (Shalom Abramowitsch) produced works and neologisms which eventually contributed to Israeli. However, it was not until the beginning of the twentieth century that the language was first spoken.

During the past half-century, Israeli has become the official language of Israel, acting as the primary mode
of communication throughout all state and local institutions and in all domains of public and private life. Yet,
with the growing diversification of Israeli society, it has come also to highlight the very absence of a unitary
civic culture among citizens who seem increasingly to share only their language. The exalted status currently
enjoyed by Israeli is, in fact, the result of an ideological process linking its historical development with the
politics of national revival.

As a result of distinctive characteristics, such as the lack of a continuous chain of native speakers from Hebrew
to Israeli, Israeli presents the linguist with a unique laboratory in which to test a wider set of theoretical
problems concerning language genesis and evolution. Indeed, the genetic classification of Israeli has
preoccupied linguists from the beginning of the twentieth century. The still prevalent, traditional view
suggests that Israeli is Semitic: (Biblical/Mishnaic) Hebrew revived (cf. Rosén 1956 and Rabin 1974).
Educators, scholars and politicians have contributed to this assumption, in their efforts to impose a
nationalist narrative on linguistic reality. Revisionists counter that it is actually Indo-European: Yiddish
‘relexified’, that is Yiddish with Hebrew vocabulary (cf. Horvath and Wexler 1997). My own hypothesis is
that Israeli is a hybrid language, both Semitic and Indo-European. I argue that both Hebrew and Yiddish
act equally as its primary contributors, accompanied by a plethora of other contributors such as Polish, Russian, German, Judaeo-Spanish (a.k.a. 'Ladino'), Arabic, English etc. Thus, the term Israeli is far more appropriate than 'Israeli Hebrew', let alone 'Modern Hebrew' or 'Hebrew' tout court.

(i) <<< Two Primary Contributors: HEBREW <-> YIDDISH
(ii) <<< Many Additional Contributors: Russian, Polish, English, German, Judaeo-Spanish ('Ladino'), Arabic etc.

This book seeks to expose as myths some of the linguistic assumptions that traditionalists
(and in some cases revisionists) take for granted:
(1) The Stammbaum Model Myth,
(2) The Internal Development Myth,
(3) The Literary as Spoken Language Myth,
(4) The Mutual Intelligibility Myth, and
(5) The Thin Language Myth.


The Stammbaum Model (family tree) insists that every language has only one parent. To refute this, some revisionists have drawn heavily on linguistic debate about creole languages. Undoubtedly, creolistics has an important bearing on the problem of Israeli. Yet, I would challenge their particular application. The terms substratum and superstratum, invoked by revisionists to support their argument that Israeli is fundamentally Yiddish, are often used in creolistics and in studies of language evolution to describe the relative influence of one language on another. Traditionally, the substratum is the base language, which determines the structural foundations of the emerging creole. The superstratum is the prestigious language influencing the emerging creole from above, especially with regard to vocabulary. Israeli revisionists contend that Yiddish, the revivalists’ mother tongue, is the substratum whilst Hebrew is only a superstratum. However, ‘substratum’ and especially ‘superstratum’ are politically charged, and often misleading.

The reality of linguistic genesis is far more complex than a simple family tree system allows. It might well be the case that ‘each language has a single parent’ ‘in the normal course of linguistic evolution’ (as emphasized by Dixon 1997: 11-13, italics mine). However, an engineered language (cf. ‘non-genetic language’ in Thomason and Kaufman 1988) or a semi-engineered language like Israeli (semi because the Yiddish contribution was not ‘planned’), can be, for example, 40% Hebrew, 40% Yiddish, 10% Polish, 10% Russian, 10% English, 5% German, 5% Arabic and 5% Judaeo-Spanish. Thus, the comparative method of reconstruction (cf. Hock 1986, Anttila 1989, McMahon 1994), as well as comparative lexico-statistics (cf. Swadesh 1952) – though useful in many cases – cannot alone explain the ‘genetics’ (the study of how languages came to be) of all languages.

A principle which weakens the Stammbaum Model, casts light on the complex genesis of Israeli, and explains why the sum of the figures above can amount to more than 100% is the Congruence Principle: ‘if a feature exists in more than one contributor – whether primary or additional– it is more likely to persist in the target language’ (see Zuckermann 2003). Thus, the Subject-Verb-Object syntax of Israeli might be based simultaneously on that of standard European and on the marked order (for emphasis/contrast) of Mishnaic Hebrew (rather than Biblical Hebrew). Interestingly, the combination of Semitic and Indo-European influences is a phenomenon which can be seen occurring already in the primary contributors to Israeli themselves. Whilst (slavonicized Germanic) Yiddish was shaped by Hebrew and Aramaic, Indo-European languages, such as Greek, played a role in (Semitic) Hebrew. In fact, before the emergence of Israeli, Yiddish itself influenced some Medieval and Modern Hebrew variants (see Glinert 1991) which then in turn influenced Israeli together with Yiddish.


The second myth, which underpins the conservative view of Israeli as pure Hebrew, is the Internal Development Myth. It argues – inter alia because of an unfortunate lack of distinction between Hebrew linguistics and Israeli linguistics – that every linguistic feature in Israeli is a result of an internal development within Hebrew. The extent to which western languages affect Israeli is controversial, not least because purists would prefer ‘Modern Hebrew’ to remain ‘uncontaminated’.

Goldenberg (1996: 151-8) claims that the changes in the phonetic/phonological system of Israeli are no different from changes observed in Hebrew. However, I believe that the phonetic system of Israeli is not, as his analysis seems to suggest, a result of internal convergence and divergence within Hebrew. Rather, it is mostly a result of employing the phonology of Yiddish (and Yiddish-like Ashkenazic Hebrew).

A creolistic tool known as the Founder Principle (cf. Harrison et al. 1988 and Mufwene 2001) may profitably be adapted to the case of Israeli. It is often used to explain why the structural features of creoles are largely predetermined by the characteristics of the languages spoken by the founder population, i.e. by the first colonists. In the context of Israeli, it could be argued that Yiddish is a primary contributor to Israeli because it was the mother tongue of the vast majority of the revivalists (e.g. Ben-Yehuda) and first pioneers in Palestine at the end of the nineteenth century and the beginning of the twentieth. All the other languages which have influenced Israeli – except for Hebrew – are additional contributors. Unlike the anti-revivalist revisionists, I would argue that Hebrew too fulfills the criteria of a primary contributor because, despite its 1700 years without native speakers, it persisted as an important literary and liturgical language throughout the generations. Thus, for example, whilst Israeli phonetics, phonology and syntax are primarily European, its morphology and basic vocabulary are mainly – albeit not exclusively – Semitic.


Some Israeli grammarians – just like many laymen – do not differentiate categorically between a literary language and a spoken mother tongue. Joshua Blau (1981) compares Israeli to ‘Modern Standard Arabic’, claiming that western European influence on Israeli is similar to western European influence on Modern Standard Arabic. He admits that Israeli is more distant from ‘Classical Hebrew’ than Modern Standard Arabic from Classical Arabic, but insists that the difference is quantitative rather than qualitative (1976: 112). However, whilst Israeli is a spoken mother tongue, Modern Standard Arabic – as opposed to the various vernacular Arabics – is not (although, intriguingly, the University of Iowa is looking for a professor with ‘native or near-native fluency in Modern Standard Arabic’, see Linguist List, 1 July 2004).

On the other hand, many linguists classify Israeli in the category of modernized Semitic vernaculars, just like Palestinian Arabic. But comparing Israeli to Semitic languages characterized by both Indo-European traits (like Israeli) and a continuous chain of native speakers (unlike Israeli) is problematic. This book demonstrates that the formation of Israeli was not a result of language contact between Hebrew and a prestigious, powerful superstratum such as English in the case of Arabic, or Kurdish in the case of Neo-Aramaic. Rather, ab initio, Israeli had two primary contributors: Yiddish and Hebrew. While Kurdish is a superstratum of Neo-Aramaic, Yiddish is a primary contributor to Israeli. The two cases are, therefore, not parallel.

Any credible answer to the enigma of Israeli requires an exhaustive study of the manifold influence of Yiddish on this ‘altneulangue’ (cf. the classic Altneuland, written by Theodor Herzl, the visionary of the Jewish State in the old-new land). The ultimate question is whether it is possible to bring an unspoken language back to life without the occurrence of cross-fertilization from the revivalists' mother tongue. At the beginning of the twentieth century, Yiddish and Hebrew were rivals to become the language of the future Jewish state. At first sight, it appears that Hebrew has won and that Yiddish after the Holocaust was destined to be spoken almost exclusively by Orthodox Jews and some eccentric academics. Yet, closer scrutiny challenges this perception. The victorious Hebrew may, after all, be partly Yiddish at heart. In other words, Yiddish survives beneath Israeli phonetics, phonology, syntax, lexis and even morphology, although traditional and institutional linguists have been reluctant to admit it.


The fourth conservative myth, the Mutual Intelligibility Myth, dictates that Israeli is Hebrew because an Israeli speaker can understand Hebrew. The linguist Edward Ullendorff has claimed that the biblical Isaiah could have understood Israeli. I am not convinced that this would have been the case. The reason Israelis can be expected to understand the book of Isaiah – albeit with difficulties – is surely because they study the Old Testament at school for eleven years, rather than because it is familiar to them from their daily conversation. Furthermore, Israelis read the bible as if it were Israeli and often therefore misunderstand it. When an Israeli reads yéled sha‘ashu‘ím in Jeremiah 31:19 (King James 20), s/he does not understand it as ‘pleasant child’ but rather as ‘playboy’. Ba’u banim ‘ad mashber in Isaiah 37:3 is interpreted by Israelis as ‘children arrived at a crisis’ rather than as ‘children arrived at the mouth of the womb, to be born’. The available examples are not only lexical: Israelis are often incapable of recognizing moods, aspects and tenses in the Bible.

Yet, Israeli children are brainwashed into believing that the Old Testament was written in their mother tongue. In other words, in Israeli primary schools, Hebrew and Israeli are, axiomatically, the very same. One cannot therefore expect Israelis easily to accept the idea that the two languages might be genetically different. In English terms, it is as if someone were to try and tell a native English-speaker that his/her mother tongue is not the same as Shakespeare’s. The only difference is that between Shakespeare and the current native speaker of English there has been a continuous chain of native speakers. By contrast, between the biblical Isaiah and contemporary Israelis there has been no such chain, while the Jews have had many mother tongues other than Hebrew.

On the other hand, even if Israelis can understand some Hebrew, that does not automatically mean that Israeli is a direct continuation of Hebrew only. Mutual intelligibility is not so crucial in determining the genetic affiliation of a language. After all, speakers of Modern English cannot understand Chaucer, but no one would claim that his language is not genetically related to contemporary English. By contrast, a Frenchman might understand Haitian Creole but who would argue that the latter is based only on French?


Israeli educators, as well as laymen, often argue that Israelis ‘slaughter’ or ‘rape’ their language by ‘lazily’ speaking ‘bad Hebrew’, full of ‘mistakes’ (e.g. http://www.lashon.exe.co.il). Most Israelis say bekitá bet rather than the puristic bekhitá bet ‘in the second grade’ (note the spirantization of the /k/ in the latter); éser shékel rather than asar-á shkal-ím ‘ten shekels’ (the latter having a polarity-of-gender agreement – with a feminine numeral and a masculine plural noun). But I believe that native speakers do not make mistakes. Educators try to impose Hebrew grammar on Israeli speech, ignoring the fact that Israeli has its own grammar, which is very different from that of Hebrew.

Thus, whereas the Hebrew phrase for ‘my grandfather’ was sav-í ‘grandfather + 1st person singular possessive’, in Israeli it is sába shel-ì ‘grandfather of me’. Similarly, whilst Hebrew often used smikhút (construct-state), in Israeli it is much less common. In a construct-state, two nouns are combined, the first being modified by the second (cf. izafet in Persian and Ottoman Turkish). Compare the Hebrew construct-state ‘em ha-yéled ‘mother the-child’ with the Israeli phrase ha-íma shel ha-yéled ‘the mother of the child’, both meaning ‘the child’s mother’. Similarly, note the position of the definite article ha in the Israeli construct-state ha-òrekh dín ‘the lawyer’ (lit. ‘the arranger of law’), as opposed to the Hebrew construct-state ‘orékh ha-dín ‘id.’. Similarly, most Israeli pupils say la-bet séfer ‘to the school’ (lit. ‘to the house book’), rather than the puristic le-vét ha-séfer.

I remember a beloved primary school teacher often lionizing the ‘right’ pronunciation of the mizrahi/Sephardi Yitzhak Navon (former Israeli President) and Eliahu Nawi (former Mayor of Be’er Sheva). In his famous song Aní vesímon vemóiz hakatán, Yossi Banay writes benaaléy shabát veková shel barét, vebeivrít yafá im áin veim khet ‘With Sabbath shoes and a beret hat, and in beautiful Hebrew with Ayin and with Het’, referring to the Semitic pharyngeals [?] and [?], which most Israelis do not pronounce but which are used, for example, by old Yemenite Jews. I believe that the Yemenite pronunciation of áin and khet is non-mainstream (cf. the charged term ‘non-standard’), exactly the opposite of what Israeli children (pronouncing [none] and kh as in Bach) are told.

The linguist Menahem Zevi Kaddari has criticized the young Israeli author Etgar Keret for using a ‘thin language’ – as opposed to Shmuel Yosef Agnon, for example. In fact, when Agnon wrote ishtó méta aláv, lit. ‘his wife died/dies on him’, he meant ‘he became a widower’ (1944, cf. 1977: 13). When Keret says so, he means ‘his wife loves him very much’. However, the main problem here is that Kaddari compares Keret to Agnon as if they wrote in two different registers within the same language. However, Keret does not write in the same language as Agnon. Whilst Agnon writes in (Mishnaic) Hebrew, which is obviously not his mother tongue (Yiddish), Keret writes authentically in his native Israeli language. Israelis are not more stupid than their ancestors. Their language is not thin and their vocabulary not poor, just different.

One can see in Kaddari’s rebuke the common phenomenon of a conservative older generation unhappy with ‘reckless’ changes to the language – cf. Aitchison (2001). However, prescriptivism in Israeli contradicts the usual model, where there is an attempt to enforce the grammar and pronunciation of an elite social group. The late linguist Haim Blanc once took his young daughter to see an Israeli production of My Fair Lady. In this version, Professor Henry Higgins teaches Eliza Doolittle how to pronounce /r/ ‘properly’, i.e. as the Hebrew alveolar trill rather than as the unique Israeli uvular approximant. The line ‘The rain in Spain stays mainly in the plain’ is translated as barád yarád bidróm sfarád haérev, lit. ‘Hail fell in southern Spain this evening’. At the end of the performance, Blanc’s daughter tellingly asked, ‘Daddy, why were they trying to teach Eliza to speak like our cleaning lady?’


The binary nature of Israeli has important theoretical implications for many branches of language science: contact linguistics, sociolinguistics, language revival/survival, linguistic genetics and typology, creolistics and mixed languages. It demonstrates that genetic affiliation – at least in the case of (semi-) engineered, ‘non-genetic’ languages – is not discrete but rather a continuous line. The comparative method and lexicostatistics, though elsewhere useful, are not here sufficient. Linguists who seek to apply the lessons of Israeli to the revival of unspoken Australian (e.g. Amery 2000), Austronesian and other languages should take warning.

Israeli affords insights into the politics not only of language, but also of linguistics. One of the practical implications is that universities, as well as Israeli secondary schools, should employ a clear-cut distinction between Israeli linguistics and Hebrew linguistics. Israeli children should not be indoctrinated to believe that they speak the language of Isaiah – unless the teacher refers to the twentieth-century Israeli polymath and visionary Isaiah Leibowitz. Although revivalists have engaged in a campaign for linguistic purity, the language they created often mirrors the very cultural differences they sought to erase. The study of Israeli offers a unique insight into the dynamics between language and culture in general and in particular into the role of language as a source of collective self-perception.


The book is divided into eight chapters. Chapter 1 provides the essential background: it describes Hebrew, introduces the term ‘Israeli’ and proposes a new periodization of Hebrew and Israeli. Chapter 2 describes the current competing views about the genesis of Israeli and suggests a new theory: hybridization. Two principles are proposed: the ‘Founder Principle’ and the ‘Congruence Principle’. Chapter 3 provides a sociolinguistic background and discusses the numerous myths surrounding the language.

The second part of the book introduces and analyses the impact of Yiddish and other European languages on Israeli in all linguistic domains: phonetics and phonology (Chapter 4), morphology (Chapter 5), syntax (Chapter 6) and lexis (Chapter 7). Semantics is discussed throughout. These chapters constitute a first step towards the writing of a new grammar of the Israeli language. Up till now, many grammars have been written but all have described Israeli as Hebrew.

Chapter 8 presents conclusions, suggests some related avenues for further research, and – most importantly – discusses the numerous theoretical implications of the current research for the study of Jewish languages, contact linguistics and historical linguistics, as well as the multiple practical implications for the teaching of linguistics and ‘Hebrew’ at schools and universities in Israel and world-wide.


There are four existing volumes which this book supplements or challenges:

(1) Harshav, Benjamin 1993. Language in Time of Revolution. Stanford (California): Stanford University Press.

(2) Horvath, Julia and Paul Wexler (eds) 1997. Relexification in Creole and Non-Creole Languages – With Special Attention to Haitian Creole, Modern Hebrew, Romani, and Rumanian (Mediterranean Language and Culture Monograph Series, vol. xiii). Wiesbaden: Otto Harrassowitz.

(3) Kuzar, Ron 2001. Hebrew and Zionism: A Discourse Analytic Cultural Study (Language, Power and Social Process 5). Berlin – New York: Mouton de Gruyter.

(4) Wexler, Paul 1990. The Schizoid Nature of Modern Hebrew: A Slavic Language in Search of a Semitic Past. Wiesbaden: Otto Harrassowitz.

Whereas Harshav’s and Kuzar’s books are invaluable for cultural studies, they do not provide a linguistic theory about the genesis of the Israeli language. In fact, my book could be considered a response to Kuzar’s as yet unanswered plea that ‘[i]n order to understand how Israeli Hebrew emerged, a fresh perspective is needed, free of revivalist preconceptions’ (p. 120). Wexler’s monograph and Horvath and Wexler’s edited collection do propose a linguistic programme which reacts against revivalism, namely ‘relexification’. However, my own theory, which is not anti-revivalist, rejects relexification and suggests a new theory of Israeli genesis: hybridization.


This book is of interest to (i) general linguists of all fields, especially those working in historical, contact, typological, evolutionary and cultural linguistics; (ii) Hebraists, Semitists, Germanists, Slavists and Indo-Europeanists; (iii) sociologists and scholars of culture, politics and identity; (iv) others interested in issues of language and society.

The book is written in a style which is accessible to readers who have no prior knowledge of Israeli. Whenever a technical term, such as ‘construct-state’ (smikhút), is mentioned, it is explained. There is minimal use of abbreviations and other alienating features.


Agnon, Shmuel Yosef 1977. pitkhéy dvarím (Introductions). Tel Aviv: Shocken.

Aitchison, Jean 2001. Language Change: Progress or Decay? (3rd edition, originally 1981). Cambridge: Cambridge University Press.

Amery, Rob 2000. Warrabarna Kaurna: Reclaiming an Australian Language. Lisse (The Netherlands): Swets & Zeitlinger.

Anttila, Raimo 1989. Historical and Comparative Linguistics. Amsterdam: John Benjamins. (2nd Edition)

Blau, Joshua 1976. tkhiát haivrít utkhiát haaravít hasifrutít: kavím makbilím umafridím (The Renaissance of Hebrew in the Light of the Renaissance of Standard Arabic). (=Texts and Studies, vol. ix). Jerusalem: The Academy of the Hebrew Language.

Blau, Joshua 1981. The Renaissance of Modern Hebrew and Modern Standard Arabic: Parallels and Differences in the Revival of Two Semitic Languages. (=Near Eastern Studies, vol. xviii). Berkeley and Los Angeles: University of California Press.

Dixon, R. M. W. 1997. The Rise and Fall of Languages. Cambridge: Cambridge University Press.

Glinert, Eliezer (Lewis) 1991. ‘lemekór haivrít hakhadashá hamedubéret: iyuním batakhbír hasamúy shel ‘lefí hatáf’ ledavíd yélin’ (On the Source of Spoken Modern Hebrew: Studies of the Covert Syntax in Lefí Hatáf by David Yellin). Leshonenu 55: 107-26.

Goldenberg, Gideon 1996. ‘haivrít kelashón shémit khayá’ (Hebrew as a Living Semitic Language). halashón haivrít behitpatkhutá uvehitkhadshutá – hartsaót lerégel mlót méa shaná leyisúd váad halashón haivrít (Evolution and Renewal: Trends in the Development of the Hebrew Language – Lectures Commemorating the 100th Anniversary of the Establishment of the Hebrew Language Council) , Jerusalem: The Israel Academy of Sciences and Humanities, 148-90.

Harrison, G. A., J. M. Tanner, D. R. Pillbeam and P. T. Baker 1988. Human Biology: An Introduction to Human Evolution, Variation, Growth, and Adaptability. Oxford: Oxford University Press.

Hock, Hans Henrich 1986. Principles of Historical Linguistics. Berlin – New York – Amsterdam: Mouton de Gruyter.

Horvath, Julia and Paul Wexler (eds) 1997. Relexification in Creole and Non-Creole Languages – With Special Attention to Haitian Creole, Modern Hebrew, Romani, and Rumanian (Mediterranean Language and Culture Monograph Series, vol. xiii). Wiesbaden: Harrassowitz.

McMahon, April M. S. 1994. Understanding Language Change. Cambridge: Cambridge University Press.

Mufwene, Salikoko S. 2001. The Ecology of Language Evolution. Cambridge: Cambridge University Press.

Rabin, Chaim 1974. ‘halashón haivrít bat yaménu’ (Contemporary Hebrew). haentsiklopédya haivrít, xxvi, 660-4. Jerusalem.

Rosén, Haiim B. 1956. haivrít shelánu: dmutá beór shitót habalshanút (Our Hebrew: Its Nature from the point of view of Linguistic Methods). Tel Aviv.

Swadesh, Morris 1952. ‘Lexico-Statistical Dating of Prehistoric Ethnic Contacts: with Special Reference to North American Indians and Eskimos’. Proceedings of the American Philosophical Society 96: 452-63.

Thomason, Sarah Grey and Terrence Kaufman 1988. Language Contact, Creolization, and Genetic Linguistics. Berkeley and Los Angeles: University of California Press.

Zuckermann, Ghil`ad 2003. Language Contact and Lexical Enrichment in Israeli Hebrew. London – New York: Palgrave Macmillan.