A Comparison between Classier Languages and Classier Script: The Case of Ancient Egyptian

Orly Goldwasser
1. Introduction
Nearly the last words I heard from Prof. Polotsky on my last meeting with him were: The young generation should go back to the old masters. There, he said, they would nd many of the ideas which were now recycled or hailed as new. It was vital to return to them, both to do them justice, and to refresh our own vision with the primacy of theirs. In this article I follow the wishes of the father of the Jerusalem School by returning to the father of modern Egyptology, Jean-Franois Champollion, whose fertility of imagination and invention gave him a degree of empathy with the past which enabled him not only to crack the code of the language, but to reach startlingly accurate conclusions as to what may be called the workings of the Egyptian mind; these conclusions still stand today, and have often been repeated, unattributed, by scholars great and small through the generations. I shall concentrate on the phenomenon of Egyptian classiers, which Champollion studied on a large scale as a phenomenon of its own right, and which is only now being revisited by scholarship. Polotsky the Egyptologist never lost sight of the special case of Egyptian within the wide-ranging landscape of general linguistics. By analysing the phenomenon of Egyptian classiers in the light of modern linguistics studies of noun categorization, I hope to follow his lead. The Egyptian hieroglyphic system incorporates the most detailed classication system known in any script of the world. Yet, it has rarely been recognized as such by most Egyptologists. The signs which play the role of classiers are generally known in Egyptology by the term determinatives. The difference in terminology is not merely technical. This terminology reects the fact that most Egyptologists do not see the determinatives as a system of classiers, i.e, that reect classes, although most of them would probably agree that some of the determinatives may sometimes play the role of classiers.

Classifier Languages and Classifier Script

17

The so-called determinatives are pictograms that are placed after the vowelless root in the Egyptian script, functioning as reading aids but carrying 1 no additional phonetic value. They mark the end of words and provide semantic information about the preceding word through their iconic meaning 2 alone. The same word may sometimes take different determinatives. A word can be followed by a single determinative, two determinatives or even more. Determinatives never stand in arbitrary relation to the word they classify. First 3 appearing during the Archaic period, this mechanism reached its peak in the 4 Middle Kingdom and later. In this article I shall try to confront two major issues: a. Considerations for assessing the determinatives as classiers. The assessment will be done via a comparison between the operation of determinatives in the hieroglyphic script and the operation of classiers in classier languages. b. Assessment of the reasons which may have impeded most students of Egyptian grammar from conceiving the determinatives as a system of graphemic classiers. 2. History of research: Champollions contribution to the study of the classsier system As in many other matters, it was Champollion who identied the special semiotic role of the pictographs that occupy the nal position in the 5 hieroglyphic word. He was also the rst to coin the term determinative. He seems to have taken deep interest in this phenomenon, as he devoted a large part of his Principes generaux to the classication and discussions of the signs he recognized and dened as dterminatifs.
In the Archaic period, the Old Kingdom, and sometimes later, classiers may play a role which stands between logogram and classier. The classiers in these cases provide through their pictorial meaning essential information which completes the phonetic information provided by the preceding pictograms (Kaplony Strukturprobleme 61; Kahl System 79). The phonetic determinatives are not dealt with in this publication, as their iconic meaning has, in most cases, to be discarded; see Goldwasser Icon, and in general Schenkel Einfhrung 47. 2 To say that a classier has meaning is to say that it indicates the perceived characteristics of the entities which it classies; in other words, classiers are linguistic correlates to perception (Allan Classiers 308). 3 Kahl System 106113. 4 A word may consist of a complete unit of information without a classier. In these cases the co-text plays the role of enhancing or pointing to the choice of the correct meaning. However, in these cases, no additional iconic information is provided for the word. Most prepositions and a limited number of nouns and verbs consistently avoid classication. A discussion of this phenomenon will appear in a forthcoming publication. 5 On the history of the term see Lefebvre Grammaire 18, n.6.
1

18

Orly Goldwasser

Unlike Champollion, Gardiner dismissed the whole phenomenon of the determinative in a short statement, in which he presents it as a mere variation of his ideogram: In several of the examples the ideogram follows one or more phonograms and ends the word. In cases such as these it is called a determinative, because it appears to determine the meaning of the foregoing sound-signs and to dene that meaning in a general way 6 (Egyptian Grammar 31). This very brief denition falls short of explaining or even faithfully describing the complex phenomenon. Champollions treatment of the phenomenon of the determinative, in fact, .7 surpasses the treatment of all his followers. He presented it from a perspective which has since unfortunately been neglected by grammarians. His departure point was an attempt at the denition of the possible semantic relations that may exist between a word and its determinative. The rst group he singles out are the dterminatifs guratifs despce. He sees this kind of determinative (repeater in our terminology, see 4.1 below) as 8 la reprsantation mme de lobject dont le mot est le signe oral. The second type of possible relation dened by Champollion is dterminatifs despce, tropique (schematic relations in our terminology; see 4.3 below). Here he suggests the option of four kinds of relations, including 9 metaphorical relations. However, an examination of his list of examples (Champollion Grammaire 7981) clearly shows that only two kinds of relations can be identied in the list: a. Synecdoche: In a synecdoche, a part stands for the whole, e.g, une tte de buf signiait un buf (Champollion Grammaire 23). b. Metonymy: On gurait, la suite du mot, limage dun objet physique en rapport plus ou moin direct avec lobjet de lide exprime par le mot ainsi dtermin (Champollion 1836: 7879). Most of the examples he cites are of wine+[WINE JARS]; irtt milk+[JAR]; this type e.g. irp mna nurse+[BREAST] or hAw day+[SUN].
Gardiner Egyptian Grammar 31. For a recent discussion of terminology, see Depuydt Hieroglyphic Script. 7 An elaborate yet somehow forgotten discussion of the determinatives in Demotic, following the approach of Champollion, appears in Brugsch 1855: 2257 8 Champollion does not see the semiotic difference between the pictorial and the phonetic. A pictorial representation of a dog does not have the same signied as the word dog. For discussions of this problem in the light of modern semiotics, see Goldwasser Icon. 9 In my studies, metaphorical relations are differentiated from metonymic relations.
6

Classifier Languages and Classifier Script

19

The third kind of determinatives to be analysed by Champollion were the dterminatifs de genre (gnriques). Dening this kind of determinative, Champollion writes: Dautres signes ajouts la n des noms crits phontiquement sont, proprement parler, des dterminatifs gnriques, puisque chacun deux se joint, pour en indiquer lacception, un nombre plus ou moins considrable de noms trs-differents dans leur signication, mais qui, tous, expriment des individus ou des objets appertenent au mme genre dtres bien que despces divers (Grammaire 82; my italics). Gardiners denition, given more than a hundred years later, of what he calls generic determinatives does not go beyond Champollions denition, nor does it add to the clarication of the phenomenon: Ideograms that serve to determine a considerable number of different words can naturally only express the kind of sense borne by these, and not their specic meaning; they are therefore called generic determinatives (Gardiner Egyptian Grammar 31). This statement concerning the generic determinatives is very open-ended and says nothing about reasons or constraints for the assemblage of a considerable number of different words under a certain determinative. Nevertheless, the generic determinative (even if so vaguely and imprecisely dened), was the only kind of determinative that was acknowledged by Gardiner to be a sort of real classier. In his long discussion of the dterminatifs gnriques, Champollion makes a clear and important differentiation between central and marginal, and takes the most inclusive and important classiers of this sort to be the subject of his study. He begins with the signs and as the rst dterminatifs to be discussed, thus correctly choosing for discussion the most frequent classiers which present most central natural categories of the Egyptian script 10 and livelihood [HIDE AND TAIL=QUADRUPED] [BIRD] [SWORM] and
TAIL]

[TREE]. He was the rst to identify the iconic meaning of the [HIDE AND sign to be la moiti infrieure dune peau de buf ou de tout autre quadrupde, and he also dened its transposed meaning (to borrow a 11 Polotskian term) to be le dterminatif gnrique de tous les noms de quadrupdes dfaut des dterminatifs guratifs (Champollion 1836: 82).
10

[SWORM] , a portmanteau word = [SNAKE+WORM], see Goldwasser Wor(l)d. Champollion

denes as classier for tous les nomes de reptiles (Champollion Grammaire 86). 11 Polotsky used this term in the grammatical sense; see Polotsky Transpositions.

20

Orly Goldwasser
12

Many years later, Erman, Gardiner, and Lefbvre, present the generic determinatives or determinatives in lists according to their iconic meanings 13 (mankind, parts of body, animals, buildings etc.). This method of mechanically assembling determinatives exclusively according to their pictorial features results in an arbitrary accumulation of all sorts of classiers of different roles and meanings. It misleads the uninitiated reader, as it represents central classiers and peripheral classiers as playing an equal role in the system. It also bypasses the elaborate semiotic shift which a pictograph in the role of classier may undergo (see below). It is only during the last decade that a few Egyptologists have turned again to the path that was pioneered by 14 Champollion. 3. Why are the determinatives classiers? In his article Classiers which deals with the classier phenomenon in many languages, Keith Allan gives the following denition for morphemes identied as classiers: (a) They occur as morphemes in surface structures under speciable conditions. (b) They have meaning, in the sense that a classier denotes some salient perceived or imputed characteristic of the entity to which an associated noun 15 refers, or may refer (Allan Classiers 285) . If we exchange the word morpheme for the word grapheme in the above citation, it would be obvious that the Egyptian determinative system easily ts the requirements of a classier system. All basic phenomena that occur in the graphemic classier system have parallels in morphemic classier systems. In the following discussion, I shall describe some of these parallel phenomena. The hieroglyphic classier system is very detailed and elaborate. Unlike some of the morphemic classier systems, the script system presents a highly motivated, transparent system of classiers that is subject to a small number of
12

Edel (1955: 2425) described the Determinative as ideograms that provides the

Begriffsklasse to which certain words may belong. He calls classiers such as and Das generelle Determinativ (das Klassenzeichen), thus coming closer to the idea of classication. 13 Erman 1928: 2325; Gardiner Egyptian Grammar 3133; Lefebvre Grammaire 2223. Gardiner calls his list a list of the more important generic determinatives, while Erman and Lefebvre use simply wichtigsten Determinative and principaux dterminatifs. 14 See Rude Graphemic Classiers, a general book about classication. In Egyptology, see, Schenkel 1974; 1976; 1997: 4547; Kahl System 107; Kammerzell Aristoteles 815; Goldwasser Icon 80107; Goldwasser Wor(l)d. 15 See also Allan Natural Language 307.

Classifier Languages and Classifier Script

21

dened constraints. The reason for this transparency may be considered to be due to a difference of age. The system presented by the script is a young system, of a life span of about 3000 years, while some of the systems represented in linguistic morphemes in different languages are probably much older. As Aikhenvald remarks, The basic assumption is that if a system is semantically transparent, and connects easily with lexical sources, it is relatively new (Aikhenvald Classiers 370). Moreover, the iconic nature of the script keeps the semantic value of the graphemic classier alive, thus slowing down or even preventing depletion, grammaticalization processes and, to a large extent, fossilization by 17 convention. 4. Comparative phenomena classier languages and Egyptian classier system 4.1 The dterminatifs guratifs despce as repeaters In this stage of our discussion, many Egyptologists may feel that the denition of linguistic classiers made by a general linguist cannot be applied to the determinative phenomenon, as the hieroglyphic script shows a large number of words that receive as determinative nothing but an iconic pictogram that merely repeats in the pictorial the meaning of the word it follows e.g., sandals or carry a child. In these kinds of examples the icon is a sort of pictorial tautology of its preceding word, and a genuine 19 classication process does not seem to take place. This phenomenon, which is very central in the hieroglyphic script and plays an important semiotic role in this system, is, in my opinion, one of the main obstacles that prevented Egyptologists from recognizing the determinatives as a classier system. Nevertheless, this very phenomenon is well known in various classier languages, and makes up part of the general system of classication. Allan
16 For a detailed denition of these constraints (speciable conditions), see Goldwasser Wor(l)d 3536. 17 It may be true that most noun classes have been established on a perceptual basis; but presumably most classication is fossilized by conventions . . . (Allan Classiers 2967). 18 This statement is incorrect from a semiotic point of view. A picture is never a repetition of a word (see Goldwasser Icon, with bibliography there). For example, in the case of sandals, the signied of the word leaves open for the speaker the kind of the sandals he wants to think of, their form or color or texture, or social signicance e.g. royal sandals. The signied of the grapheme sandal forces on the speaker/reader one specic sandal of a certain shape and sometimes color (see also Goldwasser Wor(l)d 15). 19 Class-inclusion or schematic relations, see below.

18

22

Orly Goldwasser
20

(Classiers 295) calls this kind of classiers repeaters. Repeaters can be said to have identical denotation with particular nouns. The classiers activated in the Egyptian script as dterminatifs guratifs despces, play a role similar to the morphemic classiers which are dened as repeaters. Unlike other morphemic classiers, and like the dterminatif guratif despce the morphemes activated in classier languages as repeaters do not offer 21 supplementary classicatory information to their antecedent word. It may be expected that a highly iconic system such as the hieroglyphic script would explore the repeater possibility to its limits. The repeaters in the script play in many cases a most important role, in marking a specic referent 22 for a word with several referents. The word isbt carries the meanings of throne and chair respectively. It seems that the pictorial repeater may guide the reader in this case through the semantic ambiguity to the correct signied (or referent), as the word may take the classier or . In many instances the repeater may have an additional ideological value, imposing on the reader the correct signied, an image of the world which reects the values and the choices of the literate circles (for a detailed discussion of the case of the Tsm dog, see Goldwasser Wor(l)d Chapter 5). It seems to me that once the dterminatif guratifs despce can be understood as a form of classier, there is no difculty in identifying the determinative system, as a whole, as a classier system. 4.2 The determinative as classier in a taxonomic relationship When a pictogram is activated as a classier it may provide in pictorial form only an inclusive category concept on the vertical taxonomic axis on which the word is placed.
24 23

Thus, wt

embalmer, sDmi thief and Tbw

judge (literally hearer), iTA

20 The term was coined by Hla Pe in his article A re-examination of Burmese classiers. See also Aikhenvald Classiers 361362. 21 An example from Burmese is qin ta qin one house, see Allan Classiers 292. 22 See Wb I, 132. The word isbt is possibly of Semitic origin (the earliest examples date to the

18th dynasty; see Hoch Semitic Words 3639). It may also take the [WOOD] classier. In such cases, word and classier stand in stuff/object (made of) schematic relations; For schematic relations, see below 4.3 and Goldwasser Wor(l)d 3335. 23 Many systems allow variable choice of classiers; then classiers may specify the meaning of a polysemous noun (Aikhenvald Classiers 271). For other examples of this phenomenon see below 4.6. 24 For a detailed discussion of this kind of classiers, see Goldwasser Wor(l)d.

Classifier Languages and Classifier Script sandal-maker, and XArt category are all classied into the superordinate

23 category

[HUMAN+MALE] by the icon

. The nouns mnat

wet-nurse

widow are classied into the superordinate [HUMAN+FEMALE]. The words mit cat and db

hippopotamus are classied into the superordinate category of [HIDE AND TAIL] = [QUADRUPED] by the sign the hide and tail of a leopard. Niw , which is a schematization of ostrich is classied into the

superordinate category [BIRD] by the icon of the duck , which is probably the prototypical bird, the bird par excellence for the Egyptians, and as a 25 classier stands for the representation of the general concept [BIRD]. Is tomb, iHw stable, xA
26

ofce, and SS

nest take the [HABITAT] classier. Such classiers are of the generic type (dtrminatifs de genre, generic determinative). Generic or superordinate nouns very often develop into noun classiers in classier languages as well (Aikhenvald Classiers 359, 402 and passim). Common examples in different languages are classiers for categories such as [PERSON] [HUMAN] [MAN] [WOMAN] [ANIMAL] [TREE] [WOOD] [BIRD] [FISH] [WATER] or [DRINK]. 4.3 The determinative as classier in schematic/metonymic relationships Instead of denoting a set-inclusion relationship, the classier can stand in a 27 schematic (metonymic, contiguous) relationship to the word that precedes it. The word arryt the verb na gate takes the travel is followed by the
28

[HABITAT] classier, while [BOAT] classier the The word Atwt

typical Egyptian travel vehicle, the Nile boat.

25 For a discussion of the prototype in the script, see Goldwasser Wor(l)d 2729 and passim; see also Goldwasser Determinative System. 26 The nest shows two classiers, the rst a repeater, the second a taxonomic classier. The

words horizon and netherworld also may take the [HABITAT] classier, as these two locations are conceptualized as the eternal abode of man. On this classier, see below 4.7.1 . 27 On the various kinds of schematic relations in the script, see Goldwasser Wor(l)d 3335. The [BOAT] classier in schematic relation is known since the time of the Pyramid Texts, e.g., DAi cross the sky, Pyr. 128.
28

24 bed takes the

Orly Goldwasser [WOOD] classier.

The words sSmm

warm someone and tA cook take the [FIRE] classier (brazier 30 with ame rising from it). A common metonymic relationship in the script is 31 that of a container which classies a word denoting a drinkable liquid , e.g., Hnqt beer or irp wine. Classiers in schematic relations are known in classier languages. They are dened by Aikhenvald as extensions of different kinds. Among other examples she mentions the Ngangityemerri Australian language where re is used to classify all things associated with re, such as rewood charcoal, smoke, restick (Aikhenvald Classiers 404). Allan mentions a case of metonymy in Navajo, where a classier appropriate to a bag containing nails is used instead of the classier for the nails themselves (Allan Classiers 296 [after Landar]). 4.4 The determinative as classier in metaphorical relationships A rare yet alluring feature of the Egyptian classier system is that of metaphorical classiers. Recent studies in cognitive linguistics have dened metaphorical comparisons as alternative categorization, i.e., an ad hoc grouping of the conceptual world. The metaphorical classiers, which are, in effect, forms of metaphorical comparison, t in nicely with a comprehensive view of 33 Egyptian classiers as a categorization system.
Atwt bed can also appear with a repeater, e.g., . Also brand-slaves or cattle (Faulkner Dictionary 2). Champollion (Grammaire 99) writes: Plusieurs noms communs, en rapport avec les ides feu et chaleur, reoivent pour dterminatif
30 29

32

gnrique le caractre symbolico-guratif . 31 To be distinguished from the category

[WATER]. Urine, as an undrinkable liquid, may

get the [WATER] classier (e.g., Faulkner Dictionary 69), but not a [VESSEL] classier. 32 Some of the earliest known examples of classiers, dating from the Archaic period, are of this type, i.e., various vessel classiers, which classify the liquids name. For examples, see Kahl System 83 and Altenmller lmagazin (tomb of Hesire, Third Dynasty). 33 In metaphorical comparisons, as a rule, a less prototypical member of the newly created ad hoc category is compared to a prototypical member of the category. For example, in the comparison metaphor my job is a jail a new ad hoc superordinate category of unpleasant places is created, of which jail is a prototypical example. The less prototypical member of the new category (job) is compared to the more prototypical member (jail). See Shen Schemata and Shen Metaphors.

Classifier Languages and Classifier Script

25

In the Egyptian script, animalistic metaphors are a productive domain. The 34 crocodile plays the role of vehicle in some implicit metaphors which surface only in the script. The word Hnty to be greedy, which is usually classied into the [ACTION OF FORCE] superordinate category, may

occasionally also take the crocodile as a classier creating the implicit simile greedy as a crocodile, the crocodile being a prototypical member of the ad hoc category [THOSE WHO PILLAGE WHATEVER THEY 35 CAN GET]. Other examples of animal classiers that suggest metaphorical extensions are the words qnd which describe two and Dnd types of anger angry like a monkey and angry like a bull, the rst being very noisy, but not essentially dangerous and the latter quiet yet highly menacing. Another rare but charming classier is the pictogram of a cow suckling its 36 calf. It may appear as a classier in the word Ams to show solicitude. Somewhat surprisingly for the modern observer, the mother cow is the prototypical member for the Egyptians in the ad hoc category [CAREGIVING], and not the human mother. The word mni to moor, which acquired in Egyptian an additional metaphorical meaning of DEATH=[FINAL MOORING], may take as a classier [BOAT] but also the metaphoric classier [DEATH] respectively (Gardiner Egyptian Grammar 568). This metaphor is part of the deep structure conceptual metaphor [LIFE IS A JOURNEY (ON THE NILE)] which is very dominant in Egyptian world organization (Goldwasser Icon 9799; Smoczyn ski Seeking Structure). Metaphorical classiers and metaphorical transfers are a central phenomenon in classier languages (Becker Linguistic Images; Allan Classiers 296; Lakoff Women, Fire, and Dangerous Things). Scholars are in agreement that in all languages this type of classication is highly culture37 bound (Aikhenvald Classiers 311316) .
34 In The Philosophy of Rhetoric (1936), I. A. Richards distinguishes the vehicle as the basic analogy which is used in the metaphoric comparison; see Hawkes Metaphor 61 and passim.

The word skn be greedy probably also belongs here. Crocodile autopsies in modern zoos have shown crocodiles to be remarkably voracious and undiscriminating eaters. China, wood, plastic, and other such unlikely objects have been recovered from their stomachs. 36 This pictogram can also serve as a logogram. In this case, it acts in the double role of a logogram-ideogram, and the iconic signied carries the metaphorical concept. 37 The metaphorical classiers in Aikhenvalds terminology also include classiers of the Myth and Belief type. For this type of metaphorical extension, see Lakoff Women, Fire, and Dangerous Things 94; Becker Linguistic Image. In Egyptian, see Frandsen Categorization and

35

26

Orly Goldwasser

4.5 Double and multiple classication A rather common phenomenon of the hieroglyphic system of classiers is that of double classication. In some cases the script may present two possible superordinate categories. The word Hwnt maiden, for example, is classied into the superordinate [CHILDHOOD] (represented by a male child) and the superordinate [HUMAN+FEMALE]. The word mAr wretched man, pauper takes the [INFERIORITY-EVIL] 38 superordinate classier and the classier [HUMAN+MALE]. In some cases we may nd up to four classiers, as in the case of wHa fowler; sherman (FCD: 66). The rst two classiers the duck and the sh, both integral to the activity of fowling stand in metonymic (part-whole) relation to the action of fowling and shing, and thus to the fowler himself. The third classier assigns the nature of the activity of the fowler to the superordinate action category [ACTION OF FORCE] that is 39 signied by the man holding a stick, under which the fowling action is [HUMAN+MALE], which classied. Last but not least comes the classier represents the taxonomic superordinate category of the fowler himself. This kind of compound classication suggests two schematic knowledge structures and two taxonomic set inclusion superordinate classications, scaffolding the word, as it were, on all sides and in all directions. Co-ocurrence of classiers is known in classier languages although it is not permitted in all languages. Aikhenvald remarks that languages may allow the co-ocurrence of several noun classiers within one noun phrase (Aikhenvald Classiers 81). She also suggests (after Dixon) that if two classiers co-occur, one of them must be an inherent nature classier, and the other has to refer to function/use (op. cit. 8384). However, it seems that the multiple classication phenomenon is more frequent and developed in the Egyptian script than its counterpart phenomenon in classier languages. The reason may lie in the difference between the two semiotic systems. As the script classiers are not to be pronounced, they can be accumulated one on top of the other, increasing the information around the word, without causing the noun phrase to become too cumbersome.
Goldwasser Icon 94107; For a comprehensive discussion on this type of classiers, and on the Seth classier as a Myth and belief classier, see Goldwasser Metaphor.
38 39

The so-called

bad bird classier, see below note 42. hand holding a stick.

Or by its abbreviation, the

Classifier Languages and Classifier Script 4.6 Different classiers for the same word

27

The script may often use different classiers for the same word during the same period. These cases are probably not merely an outcome of an idiosyncratic choice of a scribe, but may also involve a change of focus on the different semantic components of the word. The verb rx to know receives the classier , which categorizes a wide range of words which share the same abstract, i.e., not having a concrete referent (see below 4.7). The meaning of rx ranges from know, be able to, be aware of, to the transposed Biblical meaning of knowing a woman. When referring to the last signied, the word may take the phallus classier (Faulkner Dictionary 151152). Alternating classiers are known also in classier languages. Allan writes: It often happens that a noun may be used with different classiers, either to focus deliberately on some characteristic of its referent, or simply because the referent happens to bear characteristics that are compatible with more than one classication (Allan Classiers 295). The option to use different morphemic classiers in different contexts is highly developed in Burmese. A speaker of Burmese may use eight different classiers for the word river, depending on the universe of discourse, as Becker puts it; e.g. river one line when it appears on a map, river one connection when tying two villages, or river one sacred object in mythology (Becker Linguistic Image 113). In some cases classiers in the script may refer to social status or rank. A prominent example is the word mnfAt, which carries the meanings trained soldiers, assault troops, infantry, or soldiery (FCD: 108). At least three , or . The rst classier classiers may interchange in this case carries the most general meaning, and classies the word into the category [HUMAN+MALE]. The second classier represents the more specic prototypical icon of the category [SOLDIERY] and the third classier suggests a [HIGH RANK] or the like, which should classication into the category emphasize the status of the unit.
40

A clear case of status classiers are the , [REVERED PERSON] classiers which are used mainly after personal names (Gardiner Egyptian Grammar 447; A5052). These classiers transfer a classied personal name from the general category [MALE+HUMAN] into the better category of the revered ones.
40

A sealed papyrus scroll.

28

Orly Goldwasser

Status classiers are well known in classier languages. They are widespread in East and Southeast Asian languages and in Australian languages (Aikhenvald Classiers 82). 4.7 Vitality and productivity. Foreign, mainly Canaanite, words were constantly introduced into the Egyptian lexicon, from the Old Kingdom on. However, during the New Kingdom, shortly after the end of the Hyksos rule, we witness a sharp growth in the number of 41 Canaanite words that surface in the written repertoire. In each and every case the meaning of the loan-word is analysed by the Egyptian scribes, and is almost 42 always assigned into its correct category by a classier. An intriguing example of the productivity of the system is the case of the horse. The horse is not only a new word but a new item in the world that had 43 to be analysed. The new animal (ssmt in Egyptian a loan-word) was 44 introduced into Egypt during the Hyksos period. The exciting newcomer had a profound inuence on Egyptian life, revolutionizing warfare, 45 communications, and the economy. It became a highly prestigious cultural 46 item of elite male society. Within the framework of the script, this newcomer is immediately analysed as a [HIDE AND TAIL] member and is 47 unhesitatingly accepted into the taxonomic category, as a clear example of . A good example of a productive adoption of a foreign word into the Egyptian lexicon is the Semitic word nm to slumber, to sleep (Hoch 1994: 48 185186). Similarly to the Egyptian words sDr, aawy, and qdd which carry the
41 Hundreds of loan words are known from the Egyptian texts up to the end of the New Kingdom. Many became deeply rooted in the language and are used in Demotic and Coptic texts. For a recent comprehensive monograph on the subject, see Hoch Semitic Words. 42 In some rare cases, words are assigned to wrong categories due to a misunderstanding of the Cannanite word; see Giveon Asian Toponyms. 43 On the origin of the word, see Hofmann Fuhrwesen 4243. On the horse in Egypt, see Rommelaere Les chevaux. 44 The earliest horse bone we have comes from Tell el-Daba, Thirteenth Dynasty (circa 1650 B.C., see Boessneck 1976). 45 Different crops had to be grown with attention to the special consumption of horses, stables had to be built (see the exquisite stable complex uncovered at Piramesses; see Pusch Piramesses) and special manpower trained. 46 Probably only male elite society, although during the Amarna period women of the royal family are also represented riding chariots (Rommelaere Les chevaux, pls. 60, 66bc). Otherwise only Asiatic goddesses (Leclant Astart 1960: 167; Hadley Asherah 161164) are known to have been depicted riding horses. 47 For a large collection of examples see Mller, Appendix in: Goldwasser Wor(l)d. 48

For sDr with the

[EYE] classier, see Wb IV, 390.

Classifier Languages and Classifier Script

29

meaning sleep and lie down, the loan word nm (which enters the written repertoire already in the 18th dynasty) correctly adopts the [BED+MAN] classier and the [EYE] classier, which are typical of this category of 49 Egyptian words. 50 kAmn blind man is also a loan word from Semitic. The word consistently shows two classiers, one schematic (the taxonomic [EYE] classier) and one ([HUMAN+MALE]. Precisely like the Egyptian verb Sp be

[EYE] classier, which represents a blind, the loan-word receives the metonymic extension process typical of all Egyptian words concerned with deciency. It is always the missing element that classies the word, rather than what exists. Thus idi be deaf is classied by the [EAR] classier which is typical of all words pertaining to hearing; bald is classied by [HAIR], a classier that encompasses all types of hair, as well as activities in 51 which hair is a central feature, such as mourning. A prevailing type of classier in the Egyptian script is the schematic classier which shows stuff/object (or made of) relations with its preceding word. This type of classier is also frequently activated with loan-words. The Semitic loan-word Sbt staff, rod, consistently shows the [WOOD] classier (Hoch Semitic Words 276277), strongly suggesting a beating tool made of wood. In like manner, the Semitic loan word ks cup, goblet takes the schematic classier [METAL] (Hoch Semitic Words 338339), pointing 52 to a metal vessel, to be distinguished from the common pottery cups.
49

[BED+MAN] classier may be regarded as a repeater, while the [EYE] classier is a schematic classier of the type feature/activity; the change in the state of the eyes being a central feature of the sleeping activity.
50 Two references to blind women show the Words 320. 51

[HUMAN+FEMALE] classier, see Hoch Semitic

iAkb

mourning (Faulkner Dictionary 9) takes the schematic

(metonymic) classier [HAIR] as well as the superordinate classier on the taxonomic axis. For the missing quality type of classier in the script see Goldwasser Icon 9293.
52

Other Egyptian words for cup take the

[CUP] classier (Gardiner Egyptian Grammar [METAL] schematic classier, as well

528, W10). One example of ks cited by Hoch, shows the as the

[VESSEL] taxonomic classier (jar with handles, see Gardiner Egyptian Grammar [VESSEL].

530, W23), suggesting a secondary taxonomic superordinate classication into

30

Orly Goldwasser

Allan (Classiers 290) states that [t]he strongest evidence of semantic classication is the ability of native speakers to classify new objects consistently and easily on the basis of their observed characteristics. Aikhenvald believes that: The productivity, or vitality, of a system is measured by its ability to accept and classify new members, and reanalyse and extend the semantic range of a noun categorization device over time the Jacaltec system of noun classiers can be considered as frozen. New loan-words are simply left unclassied (Aikhenvald Classiers 313). 4.8 Abusive classication Into this class may enter mainly classiers that denote groups or individuals that are the enemy of the ofcial institutional society which prescribes the script 53 system, or of a specic author of an inscription. Instead of taking the conventional classier for human male groups, i.e. these negative groups may show the classier
ENEMY]

[HUMAM+MALE], [(SLAUGHTERED)

or [(BOUND) ENEMY]. An example for this kind of reclassication is provided by the spelling of the word ist crew, company in an inscription of the early 18th dynasty. In this inscription, the word which normally shows the [HUMAN+MALE] superordinate classier, changes its classier, when the writer refers to the company of his enemy, into the 54 [(DEAD) ENEMY] classier. Another kind of abusive classication can be found in spellings of the word XArt widow. The word is usually classied into the 55 category [HUMAN+FEMALE] (see above). The [HUMAN+FEMALE] classier is the superordinate classication of women in the script (see above 4.2). Yet at least one example from the Middle Kingdom shows the spelling (FCD: 201). Here the widow is clearly moved by the
53 It seems that the script system reects the interests and values of the ruling classes. However, a thorough study of this aspect of the script is yet to be conducted. 54 Urk IV: 67, see Gardiner Egyptian Grammar 399. The same phenomenon is repeated in the same text with the word mSa army. When referring to the Egyptian army the word is written by

the logogram

; when referring to the enemys army the logogram receives an additional

classier, the [(DEAD) ENEMY]. 55 Sometimes she may also get the [HAIR] classier. In this case, [HAIR] is a schematic classier as it must have been an important component in the rituals of mourning.

Classifier Languages and Classifier Script

31

additional classier (bad bird) into the unbecoming category of 56 [INFERIOR -EVIL]. Abusive classication is a possibility practised in many classier languages. Japanese offers the possibility of abusive classication of human beings by animal classiers. A similar role is played by diminutive classiers in many languages (Allan Classiers 296). 4.9 Semantic changes on the way: lexeme, logogram, classier 4.9.1 From logogram to classier The icon employed in the role of a classier cannot be read in its iconic (or logogramic) meaning but is transposed to another meaning (see Goldwasser Wor(l)d). B Again, it was Champollion who identied the phenomenon that when a sign is activated as a determinatif generique it undergoes a change of meaning. Different sorts of icons undergo different semiotic transpositions. , which as logogram means a man, becomes the marker of The sign the category [HUMAN+MALE] which should be close to the category [MASCULINE]. The sign with the iconic meaning a woman, when activated as a classier is transposed to the generic category 57 [HUMAN+FEMALE]. Other classiers undergo changes of meaning by metaphoric or metonymic extensions. An instructive example is the agriculturally destructive sparrow , originally carrying the meaning small when used as a logogram. When playing the role of a classier, the sign is transposed into the meanings of inferior, socially inferior, and at the end of the Old Kingdom nally clearly 58 maturing into the category [INFERIOR-EVIL] or [EVIL].
56 57

Also socially inferior; see David Linfriorit. An example from the 20th dynasty may point to a possible rare use of this classier to signal

gender. The Semitic loan-word b-r-k with a clear meaning of blessings receives the classier. If not a mere mistake, it should signal the grammatical gender of the word as [FEMININE] (Hoch Semitic Words 103104; he sees the word as a masculine nominal form, disregarding the ). For other options for marking the category [FEMININE] in the hieroglyphic script, see Goldwasser Wor(l)d 8586. The existence of the category [EVIL] marked by the bad bird was already discovered by Champollion Grammaire 102; he connected it to the fact that the sparrow was a vritable au de lagriculture gyptienne une certain poque de lanne. However, the category matured fully into this meaning only after the First Intermediate Period (e.g., HaDAwt robbery Faulkner Dictionary 164). For a study of the evolution of the classier during the Old Kingdom and the First Intermediate Period, see David Linfriorit. She connects it to the famine that seized Egypt during this period, and could have enhanced the negative role of the sparrow.
58

32

Orly Goldwasser

The papyrus roll, which originally encompassed the category belonging to papyrus or belonging to writing (Wiesmann Determinative) has acquired an extended meaning like what is written on papyrus or in words, not in reality, to the modern mind a denition of [ABSTRACT] or the like. The mechanism that operated here may be that of metonymic extension. In the case of conceptual metonymy, a certain category or domain was extended by a metonymic process which involved the transition from central cases containing concrete basic level objects (in our case, materials connected to 59 writing), metonymic extension thing that exist only in writing. A typical metonymic reading of a classier is that of the [HIDE AND 60 TAIL] classier. Iconically probably representing a leopard hide (Goldwasser Wor(l)d) the classier encompasses two main clusters: a. [HIDE] a taxonomic classier for all types of hide, leather and skin (including human skin); also as a secondary schematic classier with the relation made of [HIDE AND TAIL]. 61 b. A taxonomic classier for the animal world, excluding [BIRD] [FISH] and [REPTILE]. When activated as classier in this capacity (b), the sign obtains the transposed metonymic meaning [those who have HIDE AND TAIL]. 4.9.2 The upward mobile prototype in the script In many cases in the script we nd a prototypical member of a category acting as a superordinate classier for the whole category. In such cases, the prototype moves from its original meaning to represent the whole category (Goldwasser Wor(l)d 2729 and passim). , which is the logogram for duck, but can also One such sign is Apd assume the role of classier, representing the superordinate category [BIRD] or [WINGED ONES]. However, the signied duck should clearly be discarded when
59

follows a wide range of different fowl names, as well as orioles,

Such a complicated process is exemplied by Lakoff Women, Fire, and Dangerous Things 104105) in his study on the Japanese classier hon. The Egyptian may have understood what semiotics has constantly struggled to show in the last decades: every word, even a concrete one, is always an abstractization. For a detailed discussion of the Chapter 4.
61 60

[HIDE

AND TAIL]

classier, see Goldwasser Wor(l)d [BIRD] category starts to

By the end of the New Kingdom there are rst signs that the

move towards inclusion within the System.

[HIDE AND TAIL] category; see Goldwasser Determinative

Classifier Languages and Classifier Script

33

cranes, ostriches (niw ) and falconides. Fuzzy edges members such as mosquitoes, locusts and ies, may also occasionally be 63 classied into the [WINGED ONES] category. Already in 1836 Champollion writes: On ne doit pas considrer ce signe comme un simple dterminatif despce lorsquil sajoute tous les noms phontiques des diffrentes espces doies ou de canards connues des anciens gyptiens; tels sont, par exemple, le noms suivants . . . Mais ce caractre devient un vritable dterminatif gnrique lorsquil termine des noms doiseaux de toutes les autres espces, et il tient alors la place des caractres dterminatifs guratifs quemploie habituellement le systme dcriture hiroglyphique Il nest point inutile de faire remarquer que, dans le nom phontique du scarabe sacr , ...les signes et

ne sont que des dterminatifs gnriques, le scarabe tant considr comme appartenant au genre des volatiles (Champollion Grammaire 85 64 86). Evidence from lexical material as well as from pictorial art points to the fact that the duck was the prototypical member of the [BIRD] category in ancient Egypt. As such, , when activated as classier, becomes the pictorial representation of the image-defying superordinate [WINGED ONES] or [BIRD]. In this particular case we can trace an almost parallel process of cognitive knowledge organization occurring in the lexicon as well as in the script. Raymond Faulkner elucidated the semantic shift undergone by Apd duck from its appearance in the Pyramid Texts onwards: It therefore seems not improbable that Apd originally meant simply duck (including in this term perhaps also other small water-fowl); but in due course this word came to mean bird in general, possibly because in the undrained marshes of early Egypt ducks greatly outnumbered the other species of birds (Faulkner Duck).
E.g. bik falcon CT I 97, one version out of four; the other versions show the iconic classier of a falcon, i.e., a repeater; see also CT II 42. 63 See also Goldwasser Determinative System 5658. 64 By the denition of this complex semiotic mechanism, Champollion comes very close to the concept of a classier system. 65 By far the most frequently represented species of waterfowl in Egyptian art and hieroglyphs (Houlihan Birds 71). As life in Egypt was closely related to the Nile, it seems that in the Egyptian animal kingdom waterfowl were conceived as the most prominent type of bird. See also Wb I 9, 5; in the Belegstellen the Wrterbuch adds an emphatic note: sehr oft zu allen Zeiten.
62

65

34

Orly Goldwasser

Although, as remarked by Champollion, the word Apd never completely lost its basic-level meaning of duck or goose, its primary meaning moved from duck to bird in the lexicon, and, at about the same time materialized as a general [BIRD] classier in the script. . When activated as a Another prominent example case is the pictogram classier, it moves away from its iconic ornithological meaning which should be a falconide god or Horus the falcon god into the general meaning of 66 [GOD] or [DIVINE]. This semantic movement must have occurred, since the pictogram is activated as a classier for diverse divine beings which have clear zoomorphic manifestations, such as Sobek, the crocodile god strong anthropomorphic nature, such as Amon . , or a

As a logogram, the sign carries the meaning pr house. The hieroglyph probably represents a plan of the essence of a building with a door. When activated as classier, the sign is transposed to the superordinate meaning of [HABITAT], and may classify a stable, an ofce, as well as a lions den and a birds nest (see above 4.2). On the semantic level, the noun pr house, is also extended. Rather early on, the word pr receives an additional transposed meanings of temple, estate, or institution, when 67 combined with another noun. Nevertheless, the icon, when used as an independent logogram, never loses its original meaning of house. Other generic classiers such as or may well be cases of upward movement of prototypes. represents the prototypical action of the mouth, 68 i.e., eating, later extended to stand as classier of all [ACTIONS OF MOUTH] and [WHAT GOES INTO THE BODY]. From here the road is clear into meanings of [INSIDE THE BODY]. Thus it is nothing but surprising to nd under this classier sentiments such as anger, hate, jealousy and love, as well as the Egyptian word for perception (si3). All these feelings and abilities are conceived as dwelling in the body.
66

69

Thus, the

classier represents the most basic conceptual

See Shalomi-Hen Classifying and also the forthcoming Shalomi-Hen Writing. E.g. pr Imn Amons temple lit. the house of Amon; prHD treasury, lit. the house of silver; see Faulkner Dictionary 8990. 68 See Wiesmann Determinative. In early writings and (some later ones) of the word wnm to
67

eat, the icon

appears in a transitory semiotic status between a logogram and a classier,

e.g., (Faulkner Dictionary 62); see Schenkel wnm, and Kammerzell Varianz. For taste as the Ur sense of the child, see Goldwasser Wor(l)d 6 with note 12. 69 There are numerous examples of this metaphor in Egyptian texts of all periods, e.g. bit.f m Xt.i his ideas (qualities) are in my stomach. These are the words of Tutu describing his devotion

Classifier Languages and Classifier Script

35

metaphor that prevails in many cultures and societies [THE BODY IS A 70 CONTAINER], using the iconic representation of the most typical way of inserting of objects into the body. As words are often conceptualized as objects containing ideas (Lakoff and Johnson Metaphors 1011), many lexemes which involve speaking show this classier. This should be the explanation of the merger under one classier of verbs and words denoting eating, drinking, 71 feeling, talking and thinking that always puzzles students of hieroglyphs. The classier may represent the typical aggressive gesture that later became the classier for the broad category [FORCE], which includes many lexemes ranging from administrative power to different sorts of aggressive and 72 coercive actions. In the pictorial script, this process is remarkably productive. It may owe its productiveness and success to the ease in which the detailed pictorial lends itself to the depiction of the prototype. World representation by elect prototypes is a fundamental concept of the Egyptian culture and is constantly displayed in various manifestations of Egyptian cultural products two-and three-dimensional art, architecture, and the script system. 4.9.3 Morphemic classiers of classier languages Semantic changes The semantic change of nouns, once they become morphemic classiers, is a well attested phenomenon in classier languages. Nouns may undergo several semantic changes. Aikhenvald suggests four main processes, of which two are very active in the script: a. A noun with generic reference may become a generic classier e. g., superordinate nouns meaning man or woman often give rise to NOUN CLASSES; a noun meaning man becomes a marker for masculine, and one meaning woman is used for feminine (Aikhenvald Classiers 402). A close parallel can be identied in the Egyptian script (see 4.7.1 above with footnote 41). b. Noun with specic reference becomes a classier for a more general class of referents. Aikhenvald emphasises: The way in which a noun with specic reference can become a generic classier for the whole species is similar to the development from a
to King Akhenaten and his teachings; see Sandman Texts 76,12. For the conceptual metaphor [IDEAS ARE FOOD], see Lakoff and Johnson Metaphors 46. 70 For this conceptual metaphor, see Gibbs Poetics 203 71 See the collection of words under this classier in Gardiner Egyptian Grammar 442 (A2). This classier will be discussed in detail in a forthcoming publication. 72 Here compare Gardiner Egyptian Grammar 441, remarking the change this determinative undergoes.

36

Orly Goldwasser

prototype to its extensions. In Cayuga, the stem for car is used as a classier for all vehicles. In Mohawk the classier for fruit (-ahy-) is also 73 the word for berry (Aikhenvald Classiers 403). In this stage of my research, I can not conrm that every semiotic change occurring when a sign acquires the position of a classier is contained in the possibilities dened by Aikhenvald. A nal statement will require the close scrutiny of many classiers which have not yet been studied. 5. Conclusions The above list of parallel phenomena of Egyptian classiers in the script and of classiers in classier languages is far from being complete. However, it shows beyond any doubt that the determinatives are, in essence, classiers. One may ponder about the importance of the redenition of the determinative phenomenon as a structured classier system, asking whether it is not a case of mere change of terminology. The answer to this question is denitely negative. By understanding the determinative as classier, we move it from its alleged role as a context-bound, secondary element in the word, to the role of a classier that makes it part of a system that reects in detail the world organization of the Ancient Egyptians. It seems that the father of modern Egyptology, Jean-Franois Champollion, did not consider the dterminatif as a classier. I believe that what stood in his way, very much as it has stood in our way, was what he called dterminatifs guratifs despce, our repeaters. However, in his understanding of the determinative phenomenon, as well as in the space that he allotted to it in his grammar, he surpassed all his successors. It may indeed be that the classiers emerged in the script as a result of technical needs. They may have emerged as phonetically emptied 74 logograms, which served at the very beginning as reading aids. However, very soon this productive phenomenon gained a life of its own, making the hieroglyphic word a rich information entity, which goes far beyond the simplied task of a script, i.e reecting in writing the phonetic structure that constitutes the word. Every word in the Egyptian lexicon is represented in the script as woven into the network of world-order categorization. A word can be woven into its taxonomic axis (e.g., cup into [VESSEL]), or into a schematic axis (e.g., cup into [METAL]; see above). Due to the common phenomenon of
73 74

Cayuga and Mohawk are Iroquoian languages; see Aikhenvald Classiers 491, 497. Or sometimes partially emptied logograms.

Classifier Languages and Classifier Script

37

multiple classiers, words are often woven into the world classication map by different threads. The study of the classier system in the script is still in its infancy. I hope the coming years will yield a deeper and better understanding of this bountiful yet somewhat neglected eld of Egyptology. References
Aikhenvald Classiers A. Y. Aikhenvald, Classiers. A Typology of Noun Categorization Devices (Oxford, Oxford University Press, 2000). Allan Classiers K. Allan, Classiers , Language 53 (1977) 285311. Allan Natural Language K. Allan, Natural Language Semantics (Oxford: Blackwell, 2001). Altenmller lmagazin H. Altenmller, Das lmagazin im Grab des Hesire in Saqqara (Qs 2405), Studien zur Altgyptischen Kultur 4 (1976) 128. Becker Linguistic Image A. L. Becker, A Linguistic Image of Nature: The Burmese Numerative Classier System, Linguistics 165 (1975) 109121. Champollion Grammaire J.-F. Le Jeune Champollion, Grammaire gyptienne, principes gnraux de lcriture sacre gyptienne (Paris, Typographie de rmin Didot Frres, 1836). CT A. de Buck, The Egyptian Cofn Texts, 7 vols. Oriental Institute Publications 34, 49, 64, 67, 73, 81, 87 (Chicago, University of Chicago Press, 19351961). David Linfriorit A. David, De linfriorit la perturbation: Loiseau du mal et la catgorisation en Egypte ancienne, Gttinger Orientforschungen IV: gypten, Bd. 38. Classication and Categorization in Ancient Egypt, 1. (Wiesbaden, Harrassowitz, 2000). Depuydt Hieroglyphic Script L. Depuyt, On the Nature of the Hieroglyphic Script, Zeitschrift fr gyptische Sprache und Altertumskunde 121 (1994) 1736. Faulkner Dictionary R. O., Faulkner, A Concise Dictionary of Middle Egyptian (Oxford, Grifth Institute & Ashmolean Museum, 1962). Frandsen Categorization P. J. Frandsen, On Categorization and Metaphorical Structuring: Some Remarks on Egyptian Art and Language, Cambridge Archaeological Journal 7:1 (1997) 71104. Faulkner Duck R.O. Faulkner, Apd = duck, Journal of Egyptian Archaeology 38 (1952) 128. Gardiner Egyptian Grammar A. Gardiner, Egyptian Grammar. Being an Introduction to the Study of Hieroglyphs, 3rd edition, revised (Oxford, Grifth Institute, 1952). Gibbs Poetics R. W. Gibbs, The Poetics of the Mind. Figurative Thought, Language, and Understanding (Cambridge, Cambridge University Press, 1994). Giveon Asian Toponyms R. Giveon, Remarks on the Transmission of Egyptian Lists of Asiatic Toponyms, In: Fragen an Die Altgyptische Literature. Studien zum gedenken an E. Otto. Wiesbaden (1977) 171183. Goldwasser Icon O. Goldwasser, From Icon to Metaphor: Studies in the Semiotics of the Hieroglyphs, Orbis Biblicus et Orientalis 142 (Freibourg, Universittsverlag & Gttingen, Vandenhoeck & Ruprecht, 1995). Goldwasser Determinative System O. Goldwasser, The Determinative System as a Mirror of World Organization, Gttinger Miszellen 170 (1999) 4968 (with an appendix by M. Mller). Goldwasser Wor(l)d Classication O. Goldwasser, Lovers, Prophets and Giraffes. Wor(l)d Classication in Ancient Egypt (with an appendix by M. Mller). Gttinger Orientforschungen IV: gypten, Bd. 38. Classication and Categorization in Ancient Egypt, 3 (Wiesbaden: Harrassowitz, 2002). Goldwasser Metaphor O. Goldwasser, Where is Metaphor? Conceptual Metaphor and Alternative Classication in the Hieroglyphic Script, Metaphor & Symbol 20:2 (2005) 95 113.

38

Orly Goldwasser

Hadley Asherah J. M. Hadley, The Cult of Asherah in Ancient Israel and Judah (Cambridge: Cambridge University Press, 2000). Hawkes Metaphor T. Hawkes, Metaphor. The Critical Idiom 25 (London: Methuen, 1972). Hla Pe Re-examination Hla Pe, A re-examination of Burmese classiers, Lingua 15 (1965) 16385. Hoch Semitic Words J. E. Hoch, Semitic Words in Egyptian Texts of the New Kingdom and Third Intermediate Period (Princeton: Princeton University Press, 1994). Hofmann Fuhrwesen U. Hofmann, Fuhrwesen und Pferdehaltung im Alten gypten, Inaugural-Dissertation zur Erlangung der Doktorwrde der Philosophischen Fakultt, Rheinische Friedrich-Wilhelms-Universitt (Bonn, 1999). Houlihan Birds P. F. Houlihan, The Birds of Ancient Egypt, The Natural History of Egypt: Vol. I (Warminster: Aris & Phillips, 1986). Kahl System J. Kahl, Das System der gyptischen Hieroglyphenschrift in der 0.3. Dynastie (Wiesbaden: Harrassowitz, 1994). Kammerzell Aristoteles F. Kammerzell, Aristoteles, Derrida und gyptische Phonologie: Zu systematischen Verschiedenheiten von geschriebener und gesprochener Sprache, In Sesto Congresso Internazionale di Egittologia, II, 243251 (Torino, 1993). Kammerzell Varianz F. Kammerzell, Zur Interpretation einiger Beispiele graphemsprachlicher Varianz im gyptischen, Gttinger Beitrge zur Sprachwissenschaft, 2 (1999) 6197. Kaplony Strukturprobleme P. Kaplony, Strukturprobleme der Hieroglyphenschrift, Chronique dEgypt 41 (1966) 6099. Lakoff Women, Fire, and Dangerous Things G. Lakoff, Women, Fire, and Dangerous Things: What Categories Reveal about the Mind (Chicago & London: University of Chicago Press, 1987). Lakoff & Johnson Metaphors G. Lakoff and M. Johnson, Metaphors We Live By (Chicago, Chicago University Press, 1980). Leclant Astart J. Leclant, Astart cheval aprs les represntations gyptiennes, Syria 37 (1960) 167. Lefebvre Grammaire G. Lefebvre, Grammaire de lgyptien classique (Le Caire, Imprimerie de linstitut franias darchologie orientale, 1955). Polotsky Transpositions H.J. Polotsky, Les transpositions du verbe en gyptien classique, Israel Oriental Studies 6 (1976) 150. Pusch Piramesse E. B. Pusch, Piramesse-geliebt-von-Amun, Hauptquartier Deiner Streitwagentruppen: gypter und Hethiter in der Delta-Residenz der Ramessiden. In: Pelizus-Museum Hildesheim: Die gyptische Sammlung, edited by A. Eggebrecht, (Sonderheft der Antiken Welt, Mainz, 1993). Pyr. K. Sethe, Die altgyptischen Pyramidentexte nach den Papierabdrcken und Photographien des Berliner Museums, 3 Bde. (Leipzig, J. C. Hinrichs, 19101922). Richards Philosophy of Rhetoric I. A. Richards, The Philosophy of Rhetoric (Oxford, Oxford University Press, 1936). Rommelaere Les chevaux C Rommelaere, Les chevaux du Nouvel Empire gyptien: origines, races, harnachement (Bruxelles, Connaissance de lEgypt Ancienne, 1991). Rude Graphemic Classiers N. Rude, Graphemic Classiers in Egyptian Hieroglyphics and Mesopotamian Cuneiform, In Noun Classes and Categorization, edited by C. Craig (Amsterdam and Philadelphia, John Benjamins Publishing Company, 1986) 133138. Sandman Texts M. Sandman, Texts from the Time of Akhenaten. Bibliotheca Aegyptiaca VII, (Bruxelles, ditions de la fondation gyptologique Reine lisabeth, 1938). Schenkel Einfhrung W. Schenkel, Tbinger Einfhrung in die klassisch-gyptische Sprache und Schrift (Tbingen, im Selbstverlag des Autors, 1997). Schenkel wnm W. Schenkel, gyptisch wnm essen, Zur Interpretation der Graphien, Lingua Aegyptia 10 (2002) 5977.

Classifier Languages and Classifier Script

39

Shalomi-Hen Classifying R. Shalomi-Hen, Classifying the Divine. Determinatives and Categorisation in CT 335 and BD 17, Gttinger Orientforschungen IV: gypten. Bd. 38. Classication and Categorisation in Ancient Egypt, 2. (Wiesbaden, Harrassowitz, 2001). Shaomi-Hen Writing R. Shalomi-Hen, The Writing of Gods. The Evolution of Divine Classiers in the Old Kingdom, Gttinger Orientforschungen IV: gypten, Bd. 38. Classication and Categorization in Ancient Egypt (Wiesbaden, Harrassowitz, forthcoming). Shen Schemata Y. Shen, Schemata, Categories and Metaphor Comprehension, Poetics Today 12 (1992) 111124. Shen Metaphors Y. Shen, Metaphors and Global Conceptual Structures, Poetics 25 (1997) 117. Smoczynski Seeking Structure W. Smoczynski, Seeking Structure in the Lexicon: On Some Cognitive-Functional Aspects of Determinative Assignment, Lingua Aegyptia 6 (1999) 153162. Urk. IV K. Sethe, Urkunden der 18. Dynastie, 116. Unvernderter Nachdruck der zweiten, verbesserten Auage von 1927 und 1930 (Berlin, Akademie-Verlag, 1961) 11226. Wb A. Erman and H. Grapow, Wrterbuch der gyptischen Sprache, 7 vols. plus 5 vols. References (Leipzig, J. C. Hinrichs & Berlin, Akademie-Verlag, 19261963). Wiesmann Determinative H. Wiesmann, Die Determinative des sprechenden Mannes und der Buchrolle in den Pyramidentexten, Zeitschrift fr gyptische Sprache und Altertumskunde 57 (1922) 7378.