ON THE LANGUAGE OF LINEAR A
In a r e c e n t article1 Dr. C. H . G o r d o n claims t o have read s o m e S e m i t i c w o r d s on t h e Linear A t a b l e t s . U n f o r t u n a t e l y two of his w o r d r e c o g n i t i o n s are vitiated b y s t a r t i n g from a w r o n g r e a d i n g of t h e Linear A text, while others fail to m a t c h m e a n i n g with context2. N o r d o e s h e a t t e m p t to identify a n y g r a m m a t i c a l characteristics of t h e l a n g u a g e . Nevertheless t h o u g h t h e article m a y b e often w r o n g in detail it m a y b e right in general conclusion. Linear A m a y still r e p r e s e n t a Semitic language. T h e question is worth examination.
Preliminary.
T h e Linear A d o c u m e n t s have never b e e n a c c u r a t e l y d a t e d , b u t we m a y provisionally a s s u m e t h a t t h e y w e r e written in t h e p e r i o d 1700-1500 B. C. T h e y b e l o n g almost exclusively to C r e t e . T h e first q u e s t i o n to ask is what form of S e m i t i c l a n g u a g e could have b e e n c u r r e n t in t h a t area at t h a t t i m e . W e s t e r n Semitic, w h o s e dialects
1 Antiquity XXXI, 1957, p. 124-130.
2 It is unfortunately necessary to point out the following mistakes in Dr. Gordon's article:
8. apu identified as = «bakers», but the reading on HT 88 is adu.
9. adu sisi translated as «owner(s) of horse(s)». What he callssisi (a double L 56 which is normally considered in any case to correspond to the Linear B sign for pi), is certainly an abbreviation or ideogram rather than a phonetic word. L56 occurs doubled on HT 85a, 97a, and single on HT 27a, 89, in all cases as part of the heading of a tablet which contains the ideogram for MAN. The meaning of «horse» cannot therefore be maintained, and with it fall away the other conjectures made about the words on the tablet 85a.
12. A connection between Minoan sara2 and Semitic srr «king» is wildly unlikely. The Minoan word from its contexts must mean something like «rations»,
«issues», «rejects».
10 & 11. It is surely a mistake in method to start by assuming totally different meanings for kupa3nu and kupa3natu (the one a personal name, the other meaning «harness»).
It would be invidious to make more general criticisms, and I shall not do so.
The article however contains many suggestions which may prove fruitful.
ON THE LANGUAGE OF LINEAR A IJ
were spoken along the Mediterranean seaboard, is geographically closest to Crete. The major grammatical difference between this and the Eastern branch of Semitic is that it was the earlier to drop case- endings. But these were preserved at Ugarit in the fourteenth cen- tury, and may therefore well have been preserved by the Minoans
—assuming they were Semites— a hundred or more years earlier.
On the other hand at Alalakh Level VII and at Mari, which are the sites in the Semitic world with the closest affinities to Minoan Crete, the language used was Old Babylonian, though the personal names preserved on the tablets show that this was not the language of the bulk of the population. A similar situation may have prevailed in Minoan Crete, though if it did it would be surprising that the writing used was a linear script derived from the native picto- graphic and not cuneiform. The third alternative that must constantly be borne in mind is that any apparently Semitic words extracted from the tablets may be Babylonian loan-words.
The Minoan Language,
On the Hagia Triada tablets three syllable words are much the most common (5°°/o)i n e xt come two syllable words (32 % ) , followed by four syllable words (16 °/0). Long words are very iew. On the re- ligious material longer words are somewhat more common. Mono- syllabic words may not exist since the occasional occurrences on the tablets could be abbreviations. This distribution seems plausible enough for Semitic, the greater proportion of longer words in the more continuous texts of the sacral inscriptions being accounted for by the relevance to such texts of verbal inflections and pronominal suffixes of the sort that would find no place on the straightforward accounting tablets. But of course this is necessarily a negative argu- ment: it cannot exclude the possibility of another type of language.
Another way of checking the a priori likelihood of a Semitic language is by counting the comparative frequency of initial letters.
This procedure rests on two assumptions, first that Linear A is not a 'prefixing' language, which would badly distort any count based on initial frequencies, and second that we know the phonetic values of the signs. The only way at present that we can arrive at these is by supposing that when the Mycenaeans adopted their syllabary they used the signs to convey approximately the same sound as they had in Minoan. Fortunately there are ways of checking the accuracy
i 8 MAURICE POPE
of this a s s u m p t i o n for some of the signs1, which makes it reasonably certain t h a t this was in fact the principle on which the Mycenaeans w o r k e d . T h e following is an a p p r o x i m a t e count of H e b r e w initial frequencies compiled from Brown-Gesenius.
Aleph & weak consonants:
Labials:
k, g,Sch:
I Sir:
m : n :
q •
Sibilants:
Zayin:
Dentals:
Hebrew 26
9 16 9 6 7 4 15 3 5
Linear A
2 1
9 1-7 6 6 7 5 9
- 2 0
IOO IOO
T h e only major discrepancy between the two columns is in the sibilants and dentals. Now the later G r e e k s transliterated the first c o n s o n a n t of T y r e , which in Semitic is a palatalised sibilant s, with a dental2. S u p p o s i n g the Mycenaeans had d o n e t h e same, one of their two dental series could have stood for s in Minoan. This would not only r e m o v e the discrepancy in the table above, b u t would also explain the surprising fact that the dental series is the only one in which voiced and unvoiced stops are differentiated. T h a t this differentiation did not exist in Minoan is strongly suggested b y its well-known a b s e n c e from the derived C y p r i o t e syllabary.
In c o m p o s i n g the above table I have a s s u m e d that if Minoan were Semitic it would be their equivalent of qoph that the Mycenaeans took over as their labio-velar. A n e n c o u r a g i n g c h e c k on this is p r o v i d e d b y c o m p a r i n g the medial o c c u r r e n c e s of t h e letter:
H e b r e w (Brown-Gesenius) c. 360 initial qoph c. 175 medial Linear A 18 initial q 8 medial
1 See in particular G. Pugliese Carratelli, «La decifrazione dei testi micenei », Annuario della Scuola Archeologica Italiana di Atene XIV-XVI; Goold & Pope Preliminary Investigations, ii, x; A. Furumark, Linear A, Berlin 1956.
2 This was pointed out to me by Mr. Chadwick, who is not thereby committed to any views about the language of Linear A.
ON T H E LANGUAGE OF LINEAR A IÇ
It is worth a d d i n g that in H e b r e w qoph never follws kaph or gimel, a n d follows heth only a very few times. In Minoan q does n o t follow a k syllable in any of our existing words.
It goes without saying that the b o d y of material in Linear A is too small for the above statistics to prove a n y t h i n g positive. T h e b e s t t h e y can show is that the hypothesis that Linear A r e p r e s e n t s a Semitic language is not immediately absurd.
Grammatical Inflection.
T h e Minoan scribes naturally d o n o t tell us which of t h e i r w o r d s belong to t h e same root. Nor have t h e y left us e n o u g h material to enable us to construct a convincing table of inflections as K o b e r and Ventris were able to do with Linear B. T h e only occasions where t h e r e is a high d e g r e e of probability that we are dealing with the same word are:
kupa%nu kupa^natu kupa^natuna kupa^weja (?) dakusene dakuseneti
datare datara dataro titiku titikuni
pa3ni pannina
kupatfiu, e t c . — It is p r o b a b l e that we have here evidence of a Minoan plural form, kupagiu occurs before a singular n u m e r a l in H T 3, 49, 88, 117, 122. T h e only occasion where it p r e c e d e s a plural n u m b e r is H T I. It would s e e m t h a t on this tablet it is t h e com- m o d i t y r e p r e s e n t e d b y the h e a d i n g word that is b e i n g c o u n t e d1. T h e numeral will therefore not refer to kupa^nu. kupa^natu is found before a plural numeral in b o t h its occurrences, H T 47, a fragmen- tary tablet, and H T 119, where the list is totalled and without heading, so t h a t t h e n u m e r a l s must refer to their p r e c e d i n g words2. T h e b r o k e n libation vessel from A p o d o u l o u (Ap 2b) unfortunately gives no indication what form of the word kupaznatuna m a y b e e x p e c t e d to b e , and there can be no certainty that kupagjueja ( H T 24) is c o n n e c t e d with the s a m e root.
dakusene.— This m a y b e the singular of dakuseneti. In H T 104 dakuseneti p r e c e d e s a plural numeral: the other two words in the list,
1 See A. Furumark, Table 7.
2 See E. Peruzzi, La Parola del Passato LI, 1956, p. 437.
2 0 MAURICE P O P E
o n e of which p r e c e d e s a plural a n d t h e o t h e r a fraction, also t e r m i n a t e in -ti. In H T 103 dakusene c o m e s before a singular n u m e r a l . Unfor- t u n a t e l y t h e r e a d i n g of t h e o t h e r o c c u r r e n c e of t h e w o r d on H T 103 is u n c e r t a i n : dakusene X 6\.
T h e evidence afforded b y t h e s e two w o r d s for a Minoan plural form in -t- falls s h o r t of c e r t a i n t y . But it is n o t negligible, a n d so far as it goes it would a c c o r d well with t h e supposition of a Semitic l a n g u a g e1.
datare etc., titiku etc., pa^ni e t c . — O n t h e o t h e r h a n d I can see no w a y of explaining t h e s e w o r d s in t e r m s of Semitic g r a m m a r . titiku a n d titikun o c c u r as p e r s o n a l n a m e s at Alalakh, a n d are s u p p o - sed b y W i s e m a n to b e H u r r i a n . But if on t h e s t r e n g t h of this we b e g i n to w o n d e r if t h e language of Linear A c o u l d n o t after all b e H u r r i a n , we are i m m e d i a t e l y confronted, a m o n g o t h e r difficulties, b y t h e t e r m i n a t i o n -na. If H u r r i a n this would p r o b a b l y r e p r e s e n t t h e plural. But from the o c c u r r e n c e s of pannina, kiretana, a n d o t h e r w o r d s e n d i n g in -na t h e r e is no g r o u n d for s u p p o s i n g t h a t it indicates a plural in Linear A .
The copula.
T h e n o r m a l form of e n t r y in t h e Hagia T r i a d a lists after t h e h e a d i n g runs word — n u m e r a l , w o r d — numeral... T h e only occasions w h e r e one finds word — n u m e r a l , w o r d — w o r d — numeral, w o r d — n u - meral... are on t h e t a b l e t s 117a a n d I 2 2 a & b . On these occasions t h e s e c o n d w o r d b e g i n s with a u (L 97).
T h e discovery of t h e copula, if certain, w o u l d b e of g r e a t signi- ficance. No o t h e r c a n d i d a t e has b e e n s u g g e s t e d . But Minoan facts are h a r d to c o m e b y , and in this instance as in so m a n y o t h e r s t h e
1 Cf. Akkadian sarru «king», sarratu «queen» or «queens»; liazannu
«mayor», Txazannatu «mayors»; beliti «mistress», bcletu, beleti «mistresses».
There is of course no evidence for the gender of the words in Linear A. The main objection to kupaznatu being a plural is the occurrence of kapa%nato as a presumably Minoan name on the Linear B Knossos tablet AS1516. This cannot be conclusive. Since the word kupaznu not only inflects but occurs so frequently (in both sets of tablets and sometimes more than once on the same list) it is unlikely to be a name in Linear A. We can easily explain the fact that -it is clearly a name in Mycenaean-ruled Knossos by assuming that the word denotes a professional; cf. smith, Smith. In English Smith can be the name of a woman and Nunn the name of a man.
ON T H E LANGUAGE OF LINEAR A 21
evidence is pitifully meagre; T h e r e are however no c o n t r a d i c t o r y examples of double entries. But t h o u g h t h e A k k a d i a n for «and» is u and the H e b r e w wow used proclitically, one b e c o m e s less certain that the evidence is conclusive when one r e m e m b e r s that the copula is similar in m o d e r n Turkish and old Iranian, b o t h of t h e m non- Semitic languages1.
Suggested word meanings.
T h e search for word identifications is clearly a dangerous pastime.
A single Mycenaean sign may r e p r e s e n t m a n y p h o n e m e s , and the same m a y b e t r u e in Minoan. A t any rate all t h e possible variants must b e tested when we are looking for a word. T h e result is that the Semitic dictionaries are likely to contain at least one m a t c h for almost a n y Minoan word we try to identify. Often t h e r e are m o r e . O n e must therefore severely limit oneself to w o r d s for which the Linear A c o n t e x t is such as to give a clear indication of t h e sort of m e a n i n g to b e e x p e c t e d . T h e following list is in diminishing o r d e r of proba- bility.
kuro: T h e only Minoan word whose m e a n i n g is certainly k n o w n is kuro or kulo —«total». T h e r e is no d o u b t that this could be Semitic / A k k a d i a n kalu, kullatu; H e b r e w kôl; etc.).
kunisu: F u r u m a r k , in his excellent s t u d y of Linear A2, a r g u e d t h a t kunisu must mean some t y p e of corn3. T h e a r g u m e n t is entirely on internal evidence. It is e x t r e m e l y t e m p t i n g to c o m p a r e the frequent Babylonian kunasu, which m e a n s a species of wheat or e m m e r .
çapa3, supu, karopa$, supa3ra, pataqe are words written above various p o t s or cups on H T 3 1 . In view of the fact that the p o t s are s k e t c h e d in different ways it s e e m s m o r e likely that the w o r d s d e s c r i b e the n a m e s of the p o t s r a t h e r than their contents. Babylonian
1 In Turkish ve, in Old Iranian wa. I must thank Miss Nadia van Brock for this warning.
2 Op. cit., Text p. 21, Table 13.
3 Furumark's reasoning on kunisu seems to be both plausible in itself and to gain added credibility from Bennett's join to HT 95 (Minos III, 1955, p. 123), which he did not at the time know of. But one should not follow Furumark blindly. For instance it is hard to believe that dideru and minute in the same list should be one a personal name and the other a geographical area.
22 MAURICE POPE
parallels s u g g e s t themselves (with v a r y i n g d e g r e e s of p r o b a b i l i t y ) for four of t h e five — quppu (though this m e a n s r a t h e r a box), sappu (Ugarit sp), karpu*, saplux.
akaru: If t h e language is really S e m i t i c , akaru could h a r d l y be o t h e r than derived from the c o m m o n S e m i t i c root ' , £ / . = «eat». It o c c u r s before kunisu and corn i d e o g r a m s on H T 86a, and before c o r n i d e o g r a m s on H T 2. «Provisions» would therefore m a k e an a c c e p t a b l e h e a d i n g for b o t h tablets. T h e only objection is that on t h e l a t t e r t a b l e t t h e w o r d is parallel to kiretana, for which I can s u g g e s t n o plausible i n t e r p r e t a t i o n .
adu: T h e w o r d occurs only as a h e a d i n g , a n d can i n t r o d u c e b o t h t y p e s of t a b l e t s — the ones which consist of w o r d s and n u m e r a l s only, a n d the ones which contain c o r n i d e o g r a m s . It is frequently found t o g e t h e r with what Myres called the «contract sign» L 9 2 . All its o c c u r r e n c e s are on the tablets 85-154, which were found in t h e b u i l d i n g S . W . of the main court. It is p r o b a b l y therefore s o m e special e c o n o m i c or administrative technical t e r m which was only n e e d e d for this set of tablets2. In our p r e s e n t state of ignorance a b o u t t h e p r e c i s e rules of Minoan spelling, t h e r e are of course m a n y Semitic w o r d s we could read into it. A very plausible meaning, however, is given b y the Babylonian adu «decision», «treaty», «contract». But a similar w o r d , n o t t h o u g h t to be Semitic, occurs on the ration tablets from A l a l a k h Level VII.
I would stress that these are n o t t h e only w o r d s in t h e Minoan material for which a Semitic m e a n i n g can b e s u g g e s t e d . T h e r e are m a n y m o r e , b u t w h e r e the interpretation c a n n o t y e t b e p r e d e t e r m i - n e d from the internal evidence of the c o n t e x t s , every likeness m a y b e a m i r a g e .
General.
It d o e s not s e e m possible on t h e archaeological evidence either to refute or to s u p p o r t t h e h y p o t h e s i s t h a t t h e Minoans wrote a
1 Some of these parallels are suggested by Dr. Gordon.
2 For akaru and adu Furumark (Table 12) suggests either a geographical meaning or a general signification as a vocabulary word. He then proceeds to eliminate the latter possibility on the grounds that adu often stands next to L92 (te) —which he regards as a vocabulary word— in the same heading. But this further conclusion does not follow from his main premise, that entry words and heading words must belong to different categories.
ON T H E LANGUAGE OF LINEAR A 23
S e m i t i c l a n g u a g e , particularly as it is not n e c e s s a r y to s u p p o s e that if Linear A is Semitic, so was the Cretan p i c t o g r a p h i c . T h e new script m a y j u s t as easily have been a d a p t e d to serve t h e n e e d s of a n e w l a n g u a g e (cf. t h e adaptation of Linear B to serve Greek). In fact the a s s u m p t i o n t h a t Linear A stands to Cretan p i c t o g r a p h i c in t h e s a m e relation as E g y p t i a n hieratic to h i e r o g l y p h i c is the less likely of t h e t w o . F o r in that case we would e x p e c t religious usage to have r e t a i n e d t h e earlier script. But all the religious inscriptions in Crete are written in L i n e a r A , a n d n o t in p i c t o g r a p h i c .
T h e m o s t unequivocal piece of external evidence for the origin of t h e Minoans is of course the legend which tells us that Minos' g r a n d - father was Phoinix, or alternatively, t h e K i n g of T y r e . T h i s would suit t h e S e m i t i c h y p o t h e s i s a d m i r a b l y . But in these m a t t e r s tradition is like p r o p h e s y . It may often b e true, b u t y o u c a n n o t k n o w until i n d e p e n d e n t e v i d e n c e has p r o v e d it so.
Conclusion.
It s e e m s t o m e quite possible that s o m e of the word-identifications s u g g e s t e d a b o v e will ultimately p r o v e c o r r e c t . F o r Minoan to contain s o m e S e m i t i c w o r d s would not be surprising. Bronze A g e Crete b e l o n g e d to t h e s a m e culture as t h e c o n t e m p o r a r y N e a r E a s t . It is natural t h a t t e c h n i c a l t e r m s of a c c o u n t a n c y a n d of articles of com- m e r c e s u c h as p o t s should have b e e n b o r r o w e d . T h e further con- clusion t h a t t h e language itself is Semitic is however still a long way from b e i n g p r o v e d . T h e possible plural in -t~, t h e copula, the word for «total» in particular p r e s e n t a prima facie case. O n e c a n n o t y e t say m o r e1.
MAURICE P O P E
University of Capetotvn
1 The substance of this article formed a lecture given to the Minoan Semi- nar of the Institute of Classical Studies of the University of London and after- wards to the Linguistic Society of the University of Durham. I should like to thank the organisers of these meetings for the opportunity of addressing them and those who attended for their helpfulness in the subsequent discussions. In particular Mr. Wiseman of the British Museum has saved me from many mistakes. For those that remain I am of course solely responsible.
