Languages of the post-Diluvian World
Evolutionary theory, when applied to origins of language, fails utterly to explain the phenomena of original complexity, subsequent loss and degeneration, and the array of unrelated languages in antiquity that even now are only partially understood due to that complexity. It is here contended that only a biblical approach can explain the complicated grammar, morphology, phonetics and syntax found in ancient texts. From what we in fact find from these texts, and because these phenomena could not arise spontaneously or gradually, a supernatural interruption near the beginning of post-diluvian history is the only explanation. The supernatural interruption which created these many complex languages is precisely what is related in Genesis 11:1–9.
The origin of languages poses a major problem for evolutionists, and in the wake of Darwin’s The Origin of Species, published 1859, speculation became rife—and ridiculous—as to this matter. So outlandish were these speculations that the Société de Linguistique de Paris placed a ban on all discussion of the subject, which prevailed for more than a century.1 However, the challenge is now on for evolutionists to explain how man came to be a verbalizing creature who can communicate meaningful information through language, as Christiansen and Kirby remark:
“The recent and rapid growth in the literature on language evolution reflects its status as an important challenge for contemporary science.”2
However, this study, as well as many others, indicates that evolutionary science seeks for answers in primitive ‘symbols’, with experiments with African monkeys and other subhuman primates to ascertain (it is hoped) meaningful communication by animal gestures and signals (figure 13). Since they are committed to the view that language itself began with grunts and noises from evolutionary ape-like creatures through gestures or some kind of referral,4 they are likewise committed to the view that language evolution came concurrently with biological evolution.
Thus the simple grunts in response to, say, the presence of predators, becomes a sequence of symbols,5 which in turn moves on to simple sentences such as ‘lion in grass’ or ‘bird in tree’, and so on to ever higher and more complex arrangements of words, more complex morphology and syntax, and ultimately to abstract concepts. This ‘response-to-stimulus’ scenario is, in the immediate circumstance, the philosophy of behaviourism, which even Christiansen and Kirby reflect,6 but is really hand-in-glove with an evolutionary approach. In view of these various theories and experiments on, and observations of, non-human primates, it is noteworthy that these authors are forced to concede:
“There is inevitable scepticism regarding whether we will ever find answers to some of the questions surrounding the evolution of language and cognition.”7
There are indeed several problems with this ‘grunts-to-grammar’ evolutionary scenario. One such issue is that the further back one goes in the history of language in general, and of any language in particular, the more complex it becomes. Case endings appear at the end of nouns; prefixes, infixes, and affixes are added to verbs to modify or to expand meanings; different moods (modes) of verbal inflection occur to denote different types of expression; semantic subtleties are present to distinguish one expression from another—whether in words themselves, or word endings, or in idiomatic phrases.8
We can illustrate this with the example of Old English (that of 1,000 years ago, as spoken in Anglo-Saxon times): it had four cases for its nouns—with remnants of a fifth, each duly inflected, plus different inflections in the conjugation of verbs.9 Modern English (MnE) has largely lost inflections in nouns and pronouns (even the distinction between ‘who’ and ‘whom’ is disappearing), while in verbs it has lost the distinction between second person singular and plural, and largely lost its subjunctive mood. Moreover, Old English (OE) would distinguish parts of speech more precisely, where MnE will use the same word as a verb and as a noun or adjective.
Hence OE will distinguish the adjective open from the verb openian, where MnE has ‘open’ to denote both.10 Other differences of MnE from OE occur, for example, in strong and weak forms of the adjective,11 inflected infinitives,12 and two conjugated tenses (present and preterite), with resolved forms only beginning to appear.13 On the latter, MnE has ‘atomized’ the verb and builds tenses with either the infinitive or a participle, allied with a somewhat clumsy array of auxiliary verbs.
In summary, while Jean Aitchison concludes with a kind of ‘dynamic equilibrium’ in regard to language change, she does reject ‘language evolution’:
“Disruptive and therapeutic tendencies vie with one another in a perpetual stalemate. There is no evidence that language is evolving in any particular direction.”14
And again, citing the renowned linguist Joseph Greenberg:
“ … ‘the evolution of language as such has never been demonstrated, and the inherent equality of all languages must be maintained on present evidence’.”15
Hence this evolutionary scenario must be seriously doubted. The general tendency is in the opposite direction.
What is complexity?
Before proceeding further, some definitions of ‘complexity’ are in order, or at least the ingredients of a definition. Recent studies have attempted to analyze the problem mathematically, with little reference to actual examples of complexity or simplicity in the languages chosen.
Hence both Juola16 and Bane17 define complexity according to those aspects which can be quantified and analyzed mathematically; but without wishing to belittle these studies the question still needs to be asked, “Can language—and its complexity—be reduced to mathematics in this way?”18 Furthermore, these studies deal with contemporary languages on the plane of the present: no attempt is made to explicate any sort of historical trend to simplification.
Yet, even with their approach and criteria, it is significant that in Bane’s study, where he selects twenty languages for comparison, the most ancient language of his group, viz. Latin, comes out as the most complex, while Bislama—one of his ‘Creole’ group, appears near the bottom of his list as a very simple language, as indeed it is.19
In response to these rather abstract, not to say abstruse, exercises in quantifying and mathematical analyses of the issue, for the purposes of this study I would offer a combination of five basic criteria for complexity: economy (of words), comprehensiveness (of meaning and information), precision of expression, extent of vocabulary (including the semantic range of words and availability of synonyms), and subtleties in nuance and expression. In the last category I would put things like case endings, verbal inflections, dual and plural forms, and other such subtleties.
In illustrating the first two categories, we could cite how one word in Koine Greek, ε;λεγεν, the imperfect of λεγω, has to be translated by at least three words in English, ‘he was saying’, or four, ‘he used to say’, or even five, ‘he was going to say’.20 In particular, the combination of economy and precision is a useful guide: thus the Greek perfect of ε;ρχομαι, viz. ε;ληλυθα, ‘I have come’, is actually pregnant with meaning and difficult to translate fully without being verbose. Likewise with the Akkadian perfect of the same verb, ittalkam.21
Phonemes and phonetics
Then there are differences of sound and pronunciation: within a language these differences are properly phonemic, where a phoneme is a distinctive unit in the phonetic system of a particular language. Consider in English the words ‘pot’ and, in the reverse spelling, ‘top’: in the first example the ‘p’ is the plosive labial while the ‘t’ is the emphatic dental; then in the second the ‘t’ is the plosive dental, while ‘p’ is the emphatic labial. One way of expressing the difference is the contrastive sound unit in the mind of the speaker on one hand and the sound(s) actually pronounced as represented in the phonetic notation on the other. The latter are phonetic variations of the basic phoneme.
Furthermore, the speaker will not always be aware of the phonetic variations of the phonemes of his own language. As illustrated in the two words above, the phoneme /p/ covers for two sounds, and likewise for /t/. Now these subtle phonetic differences, present in various ancient languages in their earliest stages, will disappear in the subsequent history of those languages.
In regard to the phonemic structure of, for example, classical Semitic languages, they can indeed at times be very subtle: there are at least two different ‘t’ sounds, at least two different glottal stops, five different sibilant or ‘s-type’ sounds and so on. The exchange of one closely related sound for another within a word can alter its meaning completely.
For example, in Hebrew sar (with the letter ס, ‘samech’) is an adjective meaning ill-humoured or peevish; śar (with the letter ש, ‘śīn’, a somewhat different sound22) is a noun meaning ‘a chief, prince, or commander’. Then in Ugaritic the verb ‘ly means ‘to go up’, while a similar verb with a slightly different initial sound (but in the same laryngeal category), ģly, means ‘to lower’.
In Akkadian tebû means ‘to arise, set out’; then ţebû (emphatic ‘ț’) means ‘to sink, submerge’. However, these phonetic subtleties are often lost in the history of a language, as previously contrasting sounds or phonemes are collapsed into one phoneme.
Thus when we compare, for example, biblical Hebrew with Modern Hebrew, we soon discern that the distinction between the emphatic ‘ţ’ (ט) and normal plosive ‘t’ (ח) has largely disappeared; likewise the distinction between the glottal stops ’aleph (א) and ‘ayin (ע), and that between samech and śin.
Japhetic, Semitic, and Hamitic languages
Another problem for the evolutionist concerns the profound differences of structure between the basic language groupings. A point of interest arises here: philologists, no matter how secular, for a long time classified languages as Semitic, Hamitic, and Japhetic, after the sons of Noah. That is not to say that they believed in a Noahic Flood, or in the biblical account of Noah’s family. These designations are partly the legacy of tradition, yet, that said, secularists in the past recognized that the early history of the Ancient Near East23 reflected (for them, only in a broad sense) the dispersion of nations in three basic streams as in Genesis 10, at least as far as language classification is concerned.
Thus Japhetic referred to the Indo-European language family, Semitic to the languages of the Near East, and Hamitic to those of Egypt and Africa. However, in more recent years this threefold linguistic stream has tended to disappear, since there are more early languages than this basic scheme would indicate, as indeed is the case, since we find a plethora of unrelated languages in the very ancient world, as will be observed below. However, this should not be seen as contradictory to the biblical statements that the descendants of Japheth, Ham, and Shem spread abroad, “each according to their languages” (Genesis 10:5, 20, 31).
Greek and Latin belong to the Indo-European family; so too does Hittite, although the decipherment of the cuneiform script and its classification as Indo-European came as a shock to the system for Near Eastern scholars. Meanwhile, Linear B, since its decipherment by Michael Ventris, has been clearly recognized as an early form of Greek, a discovery which also upset the hitherto fond theories of scholars.
The still earlier Linear A presents an outstanding puzzle which no-one to date has yet cracked, although there have been several theories, ranging around whether the language could be either Semitic or Indo-European. This latter family in its earliest features is usually quite complex: case endings on nouns, and these in several declensions or sets of case endings; tense systems of verbs in multiple conjugations or verbal classes with their respective sets of verbal endings.
Another feature of this language group is its ability to run consonants together, as in English with ‘tr’ (try), ‘st’ (stand), ‘gl’ (glide), even ‘str’ (strong), and so on. Semites found this difficult, even impossible, without some sort of ‘helping vowel’. While the Hittites adopted a simplified cuneiform script from Mesopotamia to write their language—which was not a good match for this precise reason (among others)—Indo-European scripts would incorporate special characters to express this ‘double consonant’ phenomenon. Hence, for example, Greek has characters or letters to represent double sounds: ξ(‘xi’, k + s), ψ(‘psi’, p + s), ζ(‘zeta’, d + z).
However, when we move to the Semitic world of the Ancient Near East this language family has a different structure altogether. While it still has the standard parts of speech (nouns, adjectives, verbs, adverbs, etc.), the basic structure of these parts of speech derives from what is called tri-literality, i.e. that three consonants or radicals make up a root word, which is inflected with differing vowels, or modified with affixes, infixes, and suffixes to form various verbal stems, nouns, adjectives, and even prepositions. Then the way of dealing with a verb is quite different from the treatment by the Indo-European family with the latter’s tense systems, or at least quasi-tense systems.24
Semitic verbs express the quality of an action (known as an aspect system) rather than when it was done, i.e. whether the action is complete or incomplete. Concerning the Akkadian verbal system, which has a present, a preterite, a perfect, and a stative, Ungnad comments as follows:
“Originally Akkadian probably had no true ‘tenses’ in the traditional sense. Rather it distinguished actions that were punctiliar or durative in their aspects.”25
Huehnergard comments in similar fashion:
“As will be seen from the descriptions that follow … the term ‘tense’ for the Preterite, Durative, and Perfect is inadequate. None of these forms is limited to a single time value, and all involve certain aspectual notions such as (non-)duration of action and present relevance of action.”26
A similar observation applies to Hebrew, with its perfect and imperfect. To quote one grammarian:
“Each verb pattern has two aspects: a perfect and an imperfect.” To which he adds a note, “These are commonly called ‘tenses’; but ‘tense’ is a misnomer, since the perfect and imperfect do not denote time of action or state so much as type of action or state [emphasis in original].”27
The verb is modified in meaning by infixing (or prefixing) a ‘t’, or prefixing an ‘N’ to make it reflexive or passive; adding an ‘H’ or a ‘Š’ (pronounced ‘sh’) to make it causative (‘have someone do something’), or doubling the middle consonant of the normal three-consonant verb either to intensify its meaning (whereby, for example, ‘break’ becomes ‘shatter’), or again to make it causative.
Language development—complexity followed by loss
To see this pattern of original complexity, a survey of some of the main languages of far antiquity is in order here. We begin with Sumerian, the original lingua franca of Mesopotamia, at least as far as textual attestation is concerned, died out as a spoken language around the turn of the Third to the Second millennium BC. Old Assyrian and Old Babylonian, dialects of Akkadian, replaced Sumerian, but the latter remained as a classical literary language in the scribal schools, much as Latin continued for centuries in the schools and universities of Europe after the fall of the Roman Empire.
Sumerian, for its part, has interesting features. As a spoken language in the Third Millennium BC, the Sumerians developed for it the cuneiform script, later taken over by the Akkadians for writing their own language. This language is classified as agglutinative, i.e. where morphemes or units of meaning—a nominal or verbal base expressed as simple syllables—string together to form larger words equivalent to phrases, and even whole clauses and sentences (in other languages); for example, ha-ma-ab-šúm-mu means, ‘he should give it to me’.28 With the addition of prefixes, suffixes, reduplications, etc., such aggregations can become enormously complex, as for instance in the following two-word example:
me-lim5-nam-lugal-la mu4-mu4-da-zu-ne: ‘when you robe yourself with the splendour of kingship’29
When we say Sumerian is complex that is not a subjective judgment as to whether it is difficult (or easy) for person X to learn, but is so in its structure, its inflections and categories. We can see this, for example, in the Sumerian noun with its ten cases (!),30 and then its number which has at least five categories (singular, non-singular, plural, collective, detailed, etc.). Pronouns have the standard three persons, but also class (personal and non-personal), as well as number and case. And this is only the beginning (of sorrows). It is all so fiendishly complex that even now it is only about 75% understood, especially as regards the verb with its hamțu and marû stems, over which debate still continues.31 Remember that this is the earliest attested language of Mesopotamia (figure 2). Yet we are expected to believe that all this nuanced complexity had its ultimate origins in irrational grunts and noises from evolutionary brutes in response to external stimuli!
In regard to Akkadian, this Semitic language is a case in point as to the loss over time of the history of these various ancient languages. Subtleties of expression, fossilized verb forms which have lost their meaning, and distinctions in noun and pronoun forms together illustrate the pattern of loss. Hence, for example, by the end of the Old Babylonian period (conventionally 1600 BC) mimation (a final ‘m’ on singular and feminine plural nouns, pronouns, and verbal infinitives) is lost, leading to a loss of distinction between certain pronoun forms and verbal endings. Akkadian originally had a dual form as well as singular and plural, but this too died out with the passage of time. The ventive ending on verbs— conventionally understood as expressing reverse direction in verbs of motion—remained as a fossilized form while over time its meaning evaporated.32 Another loss was in initial ‘w’ words, such that wardum, a male slave, became ardu. Later on, the distinction between genitive and accusative case was lost in both singular and plural, while the infixed ‘t’ in verbs to indicate a reflexive or a passive (an infrequent form at the best of times) eventually disappeared.33
The West Semitic languages, such as Ugaritic, Phoenician, Aramaic, and Hebrew, continue, on analysis, to show various complexities and nuances, especially regarding moods and stems of verbs, which this short article cannot enumerate. Suffice it to say that what English can only express by a string of pronouns and auxiliaries with the relevant verb, a Semitic verb can express in one word, or at most two. The same goes for Akkadian.
Then there is the Egyptian of Pharaonic times, which has its own set of complications. For a start the phoneme /h/, occurs in four contrastive sounds, varying as h, ḥ, ẖ, and h̬, in order of harshness of sound, yet there is no ‘l’ sound. The verb system displays tenses (rather than aspect), as does English, but there is a past relative form to express what other languages would indicate with a relative pronoun plus a normal finite verb, and this form occurs quite commonly.34 Egyptian expresses a passive voice by participles such as ‘beloved’, ‘justified’, ‘revered’, etc.,35 but when we come to Coptic, the Egyptian of Graeco-Roman times, the passive voice has disappeared altogether—yet another example of the decline and disappearance of linguistic forms.36
Hurrian, spoken in Northern Mesopotamia and the Jezirah region (between the Upper Tigris and Middle Euphrates), has features of both agglutination and inflection. Its noun has eight cases (possibly a ninth as well), while the verb displays features similar to Sumerian (although there is no genetic relation). Hence the verb tan-, ‘to do’, will conjugate by adding first a derivative suffix to yield tan-uš-au, ‘I have done’, then adding to that to give a relative clause, tanušau-šše-ni, ‘what I have done …’. Using the verb ar-, ‘to give’, we can obtain a complex expression in a single word, ar-uš-au-šše-ni-wa-, ‘… to that which I have given’. Even now Hurrian is imperfectly understood, although a fairly extensive corpus of Hurrian texts is extant.37
Linguistic connection between Hurrian and the later Urartian was suggested early on (Sayce, 1890, etc.), and is now firmly established.38 However, Urartian, a partly agglutinative language, does have distinctive features of its own such that it is not viewed merely as a dialect or derivative of Hurrian. Hittite, sometimes known as Nesite, after the city Nesa, headquarters of the Old Assyrian colonies in Anatolia (Asia Minor in Roman times), was one of three Indo-European languages spoken on the Anatolian Plateau during the second millennium BC.39
It had a structure and vocabulary which can indeed be traced in the later languages of Europe, but had no relation at all to any of the other languages of the Ancient Near East. In Old Hittite its noun had eight cases in the singular, and up to six in the plural, with -a, -i, and -u stems, and further subdivisions for each.40 The history of the language again shows signs of disappearance of forms, and a trend to simplification: thus two of the cases in Old Hittite, the allative and the instrumental, are all but lost in Neo-Hittite. Also, the common gender nominative and accusative, distinguished in Old Hittite, merge in Neo-Hittite.41
The verb has some agglutinative features, but inflexions of vowels and consonants are also quite evident.42 There are two basic conjugations, mi- and hi-, and the verbal root can be monosyllabic or polysyllabic, with prefixes, infixes, and affixes to form a complex tense system of present and preterite, with a future understood as a variation of the present. Perfects and pluperfects are formed with the auxiliary verb har-, ‘to have’, as in certain other Western languages. In all, it is also quite complex in its own right, albeit quite different from the languages of the Semitic and Egyptian worlds. Study of the Hittite language still proceeds, and its complexities are continuing to unfold.43 Meanwhile, the Hittite language faded away with the collapse of the Hittite Empire, persisting for a while in the small states such as Carchemish, which endured after that collapse. Yet while the official language disappeared, other similar Indo-European languages persisted: Luwian, Palaic, and the languages of the Kaska and Muski peoples, who succeeded the Hittite Empire.44
Next in the discussion we consider Etruscan, the language of the inhabitants of Middle Italy prior to the coming of the Latins. In respect of archaeology their origins are obscure, albeit the story from Herodotus of a Lydian origin in the late second millennium BC has some plausibility.45 Biblically, and ultimately, it would appear that Tiras, son of Japheth, is the ancestor of the Etruscans, Genesis 10:2.46 Meanwhile, the language is fairly well understood, although it existed for several centuries before it was expressed in writing with a script adopted from the Greeks, who in turn adopted and adapted it from the Phoenicians. Again, the language is unrelated to any other, either from Europe or further afield. Of the 13,000 known inscriptions most are short epitaphs, while a few are longer, being religious texts or contracts.
In regard to structure, the Etruscan noun has inflections for each of five cases; however, there is no gender distinction in common nouns, only with proper names. The verb conjugates with the pronoun element included with the root, and adds –che to the root to form the passive. Etruscan is an economical language: unlike Hittite, it does not use auxiliaries to form perfect tenses, and adds the particle –ri to form a particular type of passive: that of obligation.47 Apart from the inevitable loan words from Greek, its vocabulary is sui generis, albeit some words passed into Latin as the latter took over as the language of the Italian Peninsula (figure 3).
Finally in this survey of very ancient languages there are the Indus Valley texts. Widely regarded as indecipherable since their discovery in the early 20th century, Barry Fells attempted a decipherment in the 1970s following methods similar to those of Michael Ventris in his work on the Linear B script. His conclusion was that the script was alphabetic, with six vowels and 24 consonants, while the language, again complex in structure, was clearly Indo-European, in turn a direct ancestor of Sanskrit.48
One could indeed go on citing examples of ancient languages as to their complexity, their subtle nuances, their economy of words to express, at times, quite extensive sentences, and in turn the gradual loss of some of these nuances over the passage of time, but the point should nevertheless be clear. What can be stated emphatically is that evolutionary theory is at a loss to explain this phenomenon of original complexity and subsequent degeneration.
The most frequent expedient is to postulate earlier ‘proto-languages’: Proto-Semitic, Proto-Sumerian, Proto-Hittite, etc. Thus Huehnergard gives a table showing East and West Semitic languages branching off from a “Common Semitic” ancestor, but then admits, “Akkadian is the earliest-attested member of the Semitic family of languages”, and again, “It is not certain when speakers of Akkadian or its linguistic predecessor(s) first arrived in Mesopotamia.”49 In other words, his “Common Semitic” is a purely hypothetical construct.
As regards any earlier form of Sumerian, Edzard admits, “The oldest reconstructable form of Sumerian dates only to about 2300 BCE, and there is a gap of at least 2,000–3,000 years between that date and the oldest comparable form of the languages under consideration”.50 This gap, however, is hypothetical, and he elsewhere offers this sobering reminder, “Our judgment on this matter is, however, highly subjective because we know nothing of the early history of Sumerian and its sound structure.”51 Yet for all that, he still talks of a ‘Proto-Sumerian’.52
When Egyptian first appears it is clearly ‘Old Egyptian’, the language spoken in the earliest phase of Egyptian history and through the Old Kingdom period. There is no ‘proto- Egyptian’. Likewise, the oldest attested Elamite text comes from the Old Akkadian period, i.e. ‘The Treaty of Naram-Sin’.53 Thus when we first encounter Elamite it is precisely that—Elamite. Speculation concerning a ‘Proto-Hurrian’ likewise remains just that—speculation, despite Wilhelm’s ‘confident assertion’.54
The same could be said for the various other ancient languages of the Fertile Crescent: there is no evidence at all that any of the postulated ‘proto-languages’ ever existed. Bonfante does not venture to talk of a ‘proto-Etruscan’, but contents himself with the simple assertion that “the Etruscans were a pocket of non-Indoeuropean speakers in an area where everyone else spoke an Indoeuropean language”.55 The one possible exception to this scenario is the Indo- European family itself (biblically the Japhetic stream): there may have been an ancestor, a ‘proto-Indoeuropean’, for Luwian, Palaic, and Hittite, but even this is conjectural. In all, they are merely theoretical constructs, born ultimately of evolutionary assumptions.
An array of disparate languages
As will have been discerned already from the above discussion another striking feature of the linguistic landscape of the Ancient Near East is the number of quite disparate and unrelated languages, but like Sumerian or Hurrian each is highly complex in its own way; several are agglutinative, like Sumerian. That they all appear on the scene at about the same time—mid-third millennium BC, and that the origins of these languages and their native speakers remain obscure is testimony to the sudden diversity and early ethnic movements of these peoples, as we would expect from Genesis 11:8–9. Some of these languages are as follows:
- Sumerian: the original language of Lower Mesopotamia, as above. It is, as seen above, an agglutinative language unrelated to any other.
- Elamite: another agglutinative language spoken in the south-western part of the Iranian Plateau. A ‘Proto- Elamite’ script (not language; figure 4) has been identified, indicating that Elamite has roots deep in the Third Millennium BC (on conventional chronology), but despite attempts to relate it to Sumerian it has no relation to any other ancient language, and is still not well understood.56
- Egyptian: from this comes Coptic, a late form of Egyptian, and the whole Hamitic or Afro-Asian family. Apart from its highly complex script (with its determinatives, ideograms, and one-, two-, and three-consonant signs),57 the language likewise is complex, but unrelated to those of the Semitic world, albeit many loan words came into those languages, e.g. Hebrew.
- Hurrian: the language of the Mitanni kingdom of the mid-second millennium BC,58 the origins of which go back to earlier times. It is first attested in cuneiform texts of the late Third Millennium BC.59 This language also has agglutinative features, and is likewise unrelated to any other ancient language.
- Hattian: the earliest language of Anatolia, of which we have only a few short texts. However, it should not be confused with the later, and unrelated, Hittite (Nesite), the Indo-European language of the Hittite Empire (see above), to which Hattian is quite unrelated. One interesting feature is the way it forms plurals, i.e. by adding a prefix, thus binu, ‘child’, becomes, in the plural, lēbinu, ‘children’.60
- Kassite: the language spoken by a people of unknown origin who overran Babylon in the period following the sack of Babylon by the Hittite king Muršilis I (conventionally c. 1595 BC). They probably came from somewhere on the western side of the Iranian Plateau or in the Zagros Mountains. This language too is unrelated to any other, and is only partly understood due to the paucity of texts.
- Semitic Family. This resolves into three further subcategories:
- East Semitic—Akkadian and its dialects. Although Akkadian has, to some extent, a common stock of vocabulary with West Semitic languages such as Hebrew, much of its vocabulary is distinct, while some derives from Sumerian. The Assyrian dialect of Akkadian is distinguished from its Babylonian counterpart mainly by differences of vowel structures and shortened forms of pronouns.61
- West Semitic: Aramaic, Hebrew, Moabite, Canaanite, Phoenician, Ugaritic, etc.
- South Semitic: Arabic, Ethiopic, Palmyrene, Nabataean.
- Etruscan: the language of the Italian residents prior to the Romans, who seem to have settled there during the Second Millennium BC. While the Etruscans adopted the Phoenician script (as modified by the Greeks) during the first half of the first millennium BC, the language itself predates this development by many centuries, and is also unrelated to any other Mediterranean or Near Eastern language.
- Indus Valley language: the script of this very early culture remains undeciphered, and the underlying language is consequently unknown—at least for those who reject the Fells decipherment. However, if Fells is right the language is Indo-European, and a direct ancestor of Sanskrit.
- Indo-European family: Hittite (Nesite), Luwian, Palaic, Sanskrit, Old Persian, Classical and Koine Greek, Latin, Old German. Hittite could well be seen as an ancestral Indo-European language.
- Uralic Group: this includes Hungarian, Finnish, Estonian, and other languages spoken around the Baltic region, and further east. They bear no relation, however, to the Slavic languages of Eastern Europe and Russia, while their origins lie in the mists of antiquity.
- Altaic Group: Turkish, Mongolian, Korean, Japanese.
- Sino-Asian Group: Tibetan, Burmese, Old Chinese.
The next observation is that all of these early languages above (up to and including the Indo-European family) are now long dead: they are no longer spoken, while several, such as Sumerian, Elamite, Hurrian, Etruscan, Kassite, and Hattian, are even now not fully understood, although for the first four we have a fair number of texts. What we can affirm here is that some of the vocabulary of these ancient languages passed into later languages, notably Hittite words which passed into Greek and Latin, and from there to languages of Western Europe. Akkadian words can also be traced in either Latin or Arabic, and via these even into some modern languages, as seen in the following samples:
- GAM.MAL: Akk. gammalu: camel, from which also we have our word, ‘camel’, the same meaning as in Sumerian.
- ÚTUL: large bowl or utility vessel; cf. Latin ūtilis: useful; French utile: useful.
muškênu: a temple dependent; cf. French mesquin: paltry, destitute.
petû: to open; cf. Latin patēre: to be open.
qarnu: horn; cf. Latin cornu: horn.
ruššu: red; cf. Latin russus: red; English russet: red.
šamaššammu: ‘oil of plant’, i.e. sunflower oil, the English
sesame derives ultimately from this Akkadian word.
e-eš-tin: to be; cf. Greek ἐστιν; Latin esse
gi-e-nu: knee; cf. Latin genu
i-ú-kán: yoke; cf. Latin iugum
kwis: who?; cf. Latin quis
The Land of Shin‘ar
Whether an occurrence in an Egyptian text of šngr or a similar occurrence in an Amarna letter of Sa-an-ha-ar refer to the biblical Shin‘ar, i.e. Lower Babylonia, is disputed, the Hebrew is nevertheless clear, even if extra-biblical references are not. According to Genesis 11:2 the confusion of languages occurred in “the land of Shin‘ar”, also mentioned in Genesis 10:10 and 14:1. That this name refers to Lower Mesopotamia is undoubted: the association with other known cities of that region in Genesis 10:10, and the destination of Shin‘ar for the Jewish exiles in Daniel 1:2, make the identification certain.
What is important is the event. It is here proposed that the confusion of languages was a supernatural act of God which created a whole array of unrelated but highly complex languages; namely, the very languages cited and discussed above, which all appear at the same time, i.e. the second half of the third millennium BC.62 Contrary to what one sometimes reads in commentaries on Genesis,63 these disparate languages were not a natural development from a single original over time, but a sudden, supernaturally induced change in the linguistic landscape with a resultant polyglot of languages. This is precisely what we find when we study ancient languages and their geographical distribution.
Language history from Babel
One final comment is necessary here: according to Genesis 10:5, 20, 31 the Japhethites, the Hamites, and the Semites spread abroad with their respective families and languages. These language groups fall into the familiar three streams: Japhetic or Indo-European, Hamitic or Afro- Asian, and Semitic—but these are not necessarily ethnic designations. One important subgroup here is, of course, the Canaanites, who populated the seaboard of the Eastern Mediterranean (Gen 10:19) and whose various subdivisions are those of the Jebusite, Amorite, etc., the ethnic groups we find in Canaan at the time of the conquest (cf. Genesis 15:19–21; Exodus 3:17; Joshua 24:11).
Now all these groups spoke variations of Canaanite, a Semitic language akin to Hebrew, which explains why the Israelites were able to converse with the Gibeonites, according to Joshua 9:6–7. However, these various peoples descended from Ham (not Shem); nevertheless they adopted Semitic languages early on as they settled in the Levant and Palestine. What is here contended is that the Tower of Babel event produced this array of disparate but highly complex languages, which remained spoken languages for several centuries, but that they eventually died out: some sooner (e.g. Sumerian and Hattian), some later (e.g. Hurrian and Etruscan). Meanwhile, others persist, or at least their linguistic descendants do: Uralic, Altaic, and Sino-Asian groups. Meanwhile, from antiquity, the main threefold language streams of Japhetic, Hamitic, and Semitic consolidated into the Indo-European, Afro-Asian, and Semitic families.
References and notes
- Christiansen, M.H. and Kirby, S., Language Evolution: consensus and controversies, Trends in Cognitive Sciences 7(7):300, 2003. However, for all the scientific discussion and citations (69 in all), this paper is highly speculative, dependent on ‘computational modelling’ (presumably with computers), appeal to “‘fossils’ of prior, more primitive stages of language” (p. 302), and exploration of various suggested ‘perspectives’. Above all this is their concern with ‘consensus’, a strange quest given the history of science, which is replete with examples of how ‘consensus’ of scientific opinion at a particular time turned out to be wrong, e.g. the phlogiston theory of combustion. Return to text.
- Christiansen and Kirby, ref. 1, p. 305. Return to text.
- Hobaiter, C. and Byrne, R., The meanings of chimpanzee gestures, Current Biology 24(14):1596–1600, 2014. Return to text.
- This itself involves the philosophy of meaning, and the ‘referral theory’ in particular—a view which is also contentious. See the discussion in Propositional semantic theories, plato.stanford.edu. Return to text.
- Christiansen and Kirby, ref. 1, pp. 301–302. Return to text.
- Note the proposal inter alia of “differentially conditioned flee responses”, Christiansen and Kirby, ref. 1, p. 301. Return to text.
- Christiansen and Kirby, ref. 1, p. 305. Return to text.
- Note here that I am confining myself to morphological and grammatical complexity, with only occasional ventures into syntactical complexity. Return to text.
- Mitchell, B. and Robinson, F.C., A Guide to Old English, 5th edn, Blackwell, Oxford, p. 62, 1992. Return to text.
- Mitchell and Robinson, ref. 9, p. 55. Return to text.
- Mitchell and Robinson, ref. 9, pp. 30–31. Return to text.
- Mitchell and Robinson, ref. 9, p. 112. Return to text.
- Mitchell and Robinson, ref. 9, pp. 36, 108. Return to text.
- Aitchison, J., Language Change: Progress or Decay? 4th edn, Cambridge University Press, New York, p. 234, 2013. Return to text.
- Aitchison, ref. 13, p. 240, citing Greenberg, J., The nature and uses of linguistic typologies, International J. American Linguistics 23(2):75, 1957. Return to text.
- Juola, P., Assessing Linguistic Complexity; in: Miestamo, M., Sinnemäki, K. and Karlsson, F. (Eds.), Language Complexity: Typology, Contact, Change, John Benjamins, Philadelphia, PA, pp. 89–108, 2008. Return to text.
- Bane, M., Quantifying and Measuring Morphological Complexity; in: Chang, C.B. and Haynie, H.J. (Eds.), Proceedings of the 26th West Coast Conference on Formal Linguistics, Somerville, pp. 69–76, 2008. Return to text.
- As for example in the formula proposed in Bane, ref. 17, p. 73. Return to text.
- Bane, ref. 17, pp. 73–74. Return to text.
- For a full discussion see Dana, H.E. and Mantey, J.R., A Manual Grammar of the Greek New Testament, Macmillan, Toronto, pp. 186–190, 1995. Return to text.
- Gperf of alākum, plus the ventive. See Huehnergard, J., A Grammar of Akkadian, Eisenbrauns, Winona Lake, p. 168, §18.1, 2005. Huehnergard is the standard grammar in this field. Return to text.
- Similar to the probable original phonetic difference in English between the ‘soft c’, as in ‘ice’, and the ‘s’, as in ‘size’, albeit the difference has long disappeared. Return to text.
- I am using the term ‘Near East’ in the strict sense as referring to the region covered by Syria-Palestine through to the Iranian Plateau. The ‘Middle East’ is properly the region from Afghanistan to Myanmar (formerly Burma), and the ‘Far East’ as comprising Thailand, Vietnam, China, Japan, etc. The use of ‘Middle East’ as referring to Palestine, Syria, and Mesopotamia, etc. is the erroneous creation of modern journalism and politics. Return to text.
- As is well known, the verbal system of Classical and Koine Greek is properly an aspect system, with the present being continuous, or incomplete action,and the aorist a completed action; but that said, it results in what is in essence a tense system. Return to text.
- Ungnad, A. (Hoffner, H.A., trans.), Akkadian Grammar, SBL Resources for Biblical Study 30, Scholars Press, Atlanta. GA, p. 62, 1992; note also the subsequent three pages as these forms are expounded. Return to text.
- Huehnergard, ref. 21, p. 17. Return to text.
- Greenberg, M., Introduction to Hebrew, Prentice-Hall, p. 45 and n. 1, 1965. Return to text.
- Edzard, D.O., Sumerian Grammar, SBL, Atlanta, GA, p. 1, 2003. Return to text.
- Edzard, ref. 28, p. 139. Return to text.
- Namely, the genitive, absolutive, ergative, dative, locative, comitative, terminative, ablative-instrumental, locative-terminative, and equative. The ergative occurs when a subject of a transitive verb exhibits a marker (i.e. the ergative case) that is different from that of the intransitive verb. See Thomsen, M.-L., The Sumerian Language, Akademisk Forlag, Copenhagen, p. 88, 1984. G. Rubio also concurs that Sumerian has ten cases, see Kaye, A.S. (Ed.), Sumerian Morphology; in: Morphologies of Asia and Africa, vol. 2, Eisenbrauns, Winona Lake, p. 1329, 2007. Return to text.
- Thomsen, ref. 30, p. 122, devotes nine pages of her grammar to this problem, then concludes with this statement, “I am well aware of the fact that the description of the hamțu and marû forms given here is rather vague. However, the system of the stems as well as that of the various conjugations of the finite verb seem to be highly inconsistent and considering the fact that Sumerian verbal forms generally are badly understood … I think that it is not possible to give a definitive answer to the question of the exact meaning and function of the hamțu and marû stems.” Return to text.
- While originally the ventive was a 1st person dative suffix (‘to me’), it occurs with verbs of motion, apparently to express a reversal of direction, but it also occurs with other types of verbs, where its lexical significance is unclear. My own theory, for what that is worth, is that it represents a change of perspective, i.e. its use depends on the perspective of the narrator—hence its use with verbs of speech. However, a full discussion of this phenomenon is outside the scope of this paper. See Ungnad, ref. 25, p. 65. See also Huehnergard, ref. 20, pp. 133–134. Return to text.
- Hence in the Kurkh Monolith Inscription of Shalmaneser III (156 lines) the Gt form occurs six times, and even there four occurrences are in fossilized forms (royal titulary, and standard, ‘template’, accounts of battle). In the Babylonian Chronicles it occurs hardly at all. Huehnergard, ref. 21, p. 390, comments that Gt forms are ‘relatively rare’. Return to text.
- Collier, M. and Manley, B., How to Read Egyptian Hieroglyphs, British Museum Press, London, pp. 68–69, 1998. Return to text.
- Collier, M. and Manley, ref. 34, pp. 98–99. A. Loprieno calls these forms ‘statives’. See Loprieno, A., Ancient Egyptian and Other Afroasiatic Languages; in: Sasson, J. (Ed.), Civilizations of the Ancient Near East (hereafter CANE), IV, Scribner’s, New York, p. 2146, 1995. Return to text.
- Lambdin, T.O., Introduction to Sahidic Coptic, Mercer University Press, Macon, GA, p. 49, 1983. Return to text.
- See Laroche, E., Glossaire de la langue Hourrite; in: Revue Hittite et Asianique, Première Partie, pp. 26, 27, 1976. Return to text.
- Wilhelm, G., The Hurrians, Aris & Phillips, Warminster, 1989. Return to text.
- Bryce, T., The Kingdom of the Hittites, Oxford, Clarendon Press, Oxford, UK, pp. 10–11, 1998. The other two are Luwian and Palaic. Return to text.
- Hoffner, H.A. and Melchert, H.C., A Grammar of the Hittite Language, Eisebrauns, Winona Lake, 2008. Return to text.
- Hoffner and Melchert, ref. 40, p. 68. Return to text.
- Hoffner and Melchert, ref. 40, pp.180–186. Return to text.
- See Held, W.H. Jr, Schmalstieg, W.R. and Gertz, J.E., Beginning Hittite, Slavica Publishers, Columbus, OH, pp.12–26, 36ff, 48, 1987. Return to text.
- Bryce, ref. 39, p. 388. Return to text.
- Fell, J., Part 2: Barry Fells’ Revolution in Deciphering Old World Scripts, 21st Century Science and Technology, Summer, p. 53, 2001. G. and L. Bonfante, cite a 6th-century BC Etruscan inscription (the ‘Warrior Stele’) on the island of Lemnos as possibly supporting the Asia Minor origin. However, whilearchaeological evidence has not so far supported the Lydian hypothesis, it remains an option. See Bonfante, G. and Bonfante, L., The Etruscan Language, Manchester University, pp. 40, 43, 51, 1983. Return to text.
- See Kidner, D., Genesis, Tyndale Press, London, p. 106, 1967. See also Leupold, H.C., Exposition of Genesis, Wartburg Press, 1942, repr. London, Evangelical Press, p. 360, 1972. Return to text.
- Bonfante, L., Reading the Past: Etruscan, British Museum Publications, London, U.K., 1990. Return to text.
- Bonfante, ref. 47, pp. 47–48. Return to text.
- Huehnergard, ref. 21, p. xxi. Return to text.
- Edzard, D.O., The Sumerian Language, in CANE IV:2107, 1995. Return to text.
- Edzard, ref. 28 , p. 4. Return to text.
- Edzard, ref. 28, pp. 3, 174. Return to text.
- Thus Gragg, G.B., Less-Understood Languages of Ancient Western Asia, in CANE IV:2165, 1995. However, Jacob Dahl has proposed a Proto-Elamite script in distinction from the linear Elamite script, which he maintains is not a true writing system. However, while he is concerned with the script there may be implications for the language, but this is unconfirmed. See Dahl, J., Early Writing in Iran: A Reappraisal”, Iran 47:32–31, 2009. Return to text.
- Wilhelm, ref. 38, p. 4. Return to text.
- Bonfante, ref. 47, p. 43. Return to text.
- See discussion by Gragg, ref. 36, pp. 2162–2163. Return to text.
- In Gardiner’s sign list for Middle Egyptian he lists, in all, 734 signs as “the commonest hieroglyphs found in Middle Egyptian”. See Gardiner, A., Egyptian Grammar, 3rd edn, Ashmolean Museum, Oxford, pp. 438, 443–542, 1957. Return to text.
- Gurney proposes that while Hurrian was the language of the populace of Mitanni, the empire was ruled by “a caste of Indo-Aryans”, whose language had affinities with Sanskrit. Gurney, O.R,. The Hittites, Penguin Books, London, UK, p. 107, 1990. Return to text.
- Edzard, ref. 28, p. 4; Wilhelm, ref. 38, p. 7. Return to text.
- Gurney, ref. 58, p. 101. See also Gragg, ref. 53, pp. 2174–2176. Return to text.
- See Huehnergard, ref. 21, pp. 599–603. Return to text.
- Care must be taken to distinguish the written attestation of a language from its actual existence as a lingua franca, which can be a gap of centuries. This is particularly so with Etruscan, and possibly Elamite (although on the latter see ref. 19). Return to text.
- As discussed in Aalders, G.Ch., Genesis, vol. I, Bible Student’s Commentary Series, Engl. Tr. Zondervan Grand Rapids, MI, pp. 253–254, 1981. Return to text.