Use the following persistent identifier: http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Benveniste.Indo-European_Language_and_Society.1973.
Chapter 3. The Sacrifice
The absence of any common term to designate the “sacrifice” is contrasted, in the separate languages and often within one and the same language, by a great diversity of expressions corresponding to the various forms which the sacrificial act may take: libation (Skt. juhoti, Gr. spéndō), a solemn verbal undertaking (Lat. voveo, Gr. eúkhomai), a sumptuous banquet (daps), fumigation (Gr. thúō), a rite of illumination (Lat. lustro).
In so far as hágios may be related to Skt. yaj-, this implies a connection between the “sacrifice” and the notion of the “sacred.” In Vedic yaj– is strictly “to sacrifice,” but first (and this is implied by the construction of the verb, accusative of the name of the god and instrumental of the object sacrificed), it meant “to honor the god, to solicit his favor, to recognize his power by means of offerings” (see above).
With this we are introduced to the study of the positive acts and the ceremonies by which the sacred is defined and maintained: these are the offerings, which are certainly “sacrifices,” means of making sacred, of transferring what is human to the divine.
These offerings take various forms and they are denoted by different terms according to whether they designate things or prayers. For the prayer is itself a kind of offering, and it acts by its effective power; in the shape of fixed formulas which accompany the rites it puts man and god in relationship to one another through the agency of the king or the priest.
The material offering may be solid or liquid: either a libation or what might be called “mactation.” It appears that the most generally attested of all the terms referring to sacrifice is that which denotes the libation. It is derived from the root which is represented in Sanskrit by hav-, juhoti ‘to offer sacrifice’, hotar– ‘sacrificial priest’, hotra– ‘sacrifice’. The corresponding Iranian form zav– also provides zaotar ‘priest’ and zaoθra– ‘sacrifice’. Here we have terms of great importance each of which is the source of numerous and frequent derivatives.
The root is also attested in Armenian by jawnem ‘offer, consecrate’ with a religious application. Finally we have the Greek khéō ‘to pour’ discussed in the previous chapter. All these forms, as we have already said, go back to the Indo-European *g’heu-, as do the present stems with enlargement, Latin fundo, Gothic giutan, ‘to pour’. This root has, therefore, in the majority of the Indo-European languages taken on a religious sense which is also shown by certain derivatives of khéō.
With reference to the “libation,” the proper sense of *g’heu– is “to pour in the fire.” In Vedic it is the liquid offering, consisting of melted butter, fat which feeds the fire and nourishes the divinity.
In this connection we may briefly recall what has been discussed above, namely a more limited correspondence which also concerns the “libation” with an interesting dialectal distribution: Gr. spéndō, spondḗ, ‘libation’, Latin spondeo, which preserves only the purpose of the act which the libation supports, namely the “engagement,” Hittite šipant– (išpant-) ‘offer a libation’ (cf. Book Six, Chapter Two, of this work).
In the Latin terminology of the sacrifice there is one word which is confined to Latin but which may be the relic of a pre-dialectal form: this is the verb mactare, the most frequent sense of which in the classical period was “to sacrifice an animal.” This cannot be separated from the nominal form mactus. This is known only in the vocative form macte, especially in the expression macte (animo) ‘courage!’, a sense which is difficult to fit in with the meaning of the verb mactare. The connection between these forms is so obscure that scholars have supposed that there are two verbs mactare, one meaning “to kill” and the other “to exalt” or something of the sort. This is an idea which is certainly to be rejected.
Mactare is to be regarded as the denominative verb from mactus, but the relation of meaning can only be elucidated by a close study of the uses. The Romans explained mactus as “magis auctus.” The literal form of this proposal cannot of course be maintained but it may be right in its basic idea, namely that of an enhancement, a reinforcement of the god, achieved by means of the sacrifice which nourishes him. It is beyond doubt that this “popular etymology” affected the use of the word macte; macte (animo) ‘be of good courage’, where macte may be explained by the sense attributed to mactus. This adjective may simply be a verbal adjective *mag-to– parallel with *mag–no– (Lat. magnus). It would not be surprising if we had two forms of the verbal adjective, one in –to– and the other in –no-; this is the case with the root *plē– from which we have both plē–nus and –plē–tus; one of these, the one in –no– indicates the natural state and the other in –to-, the state into which a thing has been changed. Thus the present denominative mactare would denote “to make big, to increase”; this is the operation which puts something in the state mactus. The oldest use mactare deum extis shows the name of the god in the accusative and the name of the sacrifice in the instrumental. It is, therefore, to make the god bigger, to exalt him, and at the same time to increase his strength by the offering. Then, by a change of construction analogous to that known from sacrare, the expression mactare victimam was coined “to offer a victim in sacrifice.” By a further development we have mactare ‘put to death, slaughter’ which is preserved in the Spanish matar ‘to kill’.
Each of these terms adds something to the idea of the sacrifice, of the offering, and the libation by the connection it establishes between the fundamental notion and the varied implications of the terms used.
Here is another example: Lat. voveo, votum certainly means “to vow, consecrate by a sacrifice,” but the correspondents of the Latin verb throw more light on the original meaning. First we have the verbal adjective in Vedic vāghat ‘making a vow of sacrifice’ and ‘sacrificing’; then Greek eúkhomai, eukhḗ. In these Greek words at first sight we seem to have a very different notion: “to pray,” “to promise” and also “to boast,” “to affirm in a solemn manner.” Finally a fourth important term of the same series is the Avestan verbal form aogǝdā ‘he said’ (3rd pers. sing, of the preterite).
We thus have a great variety of senses, one which is very precise in the Latin voveo ‘to vow’ and rather vague in the Avestan aogǝdā ‘he said’. Greek introduces a notion which is neither “to say” nor “to offer” nor “to sacrifice” but “to make a vow,” “to make a public announcement of an obligation,” “to affirm the quality of something” and consequently “to give oneself out as.” It is a solemn declaration that one pledges something or pledges oneself to do or to be something. This delimitation of sense evokes another. The verbal form of the Avestan aogǝdā is more instructive than it appears. If we take note of its uses, we see that it appears in solemn circumstances, with reference to important persons and divinities. It is a declaration which has the appearance of a promise, an undertaking, and has its authority from those who enunciate it.
We thus see that the senses have an unequal distribution in the correspondences which comprise several forms from the same root. It is not a rare occurrence that the properly religious sense is established in only one language, while elsewhere the word becomes part of the common vocabulary, or else is specialized in a different way. This remark may be illustrated by a new example, a word which has a religious sense in only one language although it enters into the lexicon of several others. This is a name for the offering which is peculiar to Latin: daps or more commonly the plural dapes, which denotes the ritual meal offered after the sacrifice. This was a term which soon was drained of its religious sense and came to denote no more than “meal.”
Here, too, although there are certain congeners, the sense to be deduced from the comparisons is still not clearly established. Along with daps we must list certain forms which deviate from it in meaning. Festus (P.F. 59, 21) defines daps as follows: “Apud antiques dicebatur res divina, quae fiebat aut hiberna sementi, aut verna.” The offering thus took place at sowing time either in the winter or the spring. Besides daps we have dapatice, adds Festus, the sense of which is “magnifiée”; dapaticum negotium, that is “amplum ac magnificum.” How can we reconcile the notion of “ample, magnificent, liberal” with that of “ritual meal”?
According to the dictionary of Ernout-Meillet the primary sense of daps was “sacrifice.” This opinion is supported by Gaius Inst. 4, 28: pecuniam acceptant in dapem, id est in sacrificium impendere ‘to spend money received for a daps, that is, for a sacrifice’. Hence comes the sense, according to E-M., “ritual meal which follows the sacrifice,” then, in the secular sense “meal, food.”
Outside Latin we have a group of words consisting of Armenian tawn ‘feast’, OIcel. tafn ‘sacrificial animal’, ‘animal destined for sacrifice’ and Greek dapánē ‘expenditure’, which is connected with dáptō ‘to divide, rend’.
This correspondence leads on to another Latin word belonging to a family and a meaning which are apparently very different: this is damnum ‘damage’, an essential term in ancient Roman law. The form damnum goes back via an ancient *dap–nom to the same type of formation as Gr. dapánē and presents the root with the same suffix –n-. But “meal,” “offering,” “expenditure,” “damage” lack any obvious unity and even seem contradictory. Consequently the Latin etymological dictionary is hesitant about admitting a connection of daps with damnum.
In our opinion the formal resemblance is sufficiently precise to warrant a search for the conditions which will make a semantic equation possible. For this it will be necessary to delimit the senses. Why should daps be a “meal” in particular and not an offering or a sacrifice; why does the derivative, the adjective dapaticus, imply lavishness and sumptuousness? Finally, how can we justify a connection, which is suggested by the form, with dapánē and also with damnum?
In our opinion it would seem that daps is not properly an offering in general to the gods but the meal offered after a consecration, a lavish and sumptuous meal. We know this type of meal in very different societies in which the point is to make an ostentatious expenditure of money. It is a “sacrifice” in the sense in which the word is used today in a spirit of parsimony: to spend money as an ostentatious act without regard for what it costs and in the knowledge that it will never be seen again. It is this attitude which is properly signified by “expenditure,” the money which is poured out for a “sacrifice” without reckoning on any return whatsoever. In much the same way in commerce the expression to sell “at a sacrificial price” is used.
Nor is it an accident that we say today (in French) “offrir un repas, un banquet” just as “offrir un sacrifice.” Daps would thus be the feast dedicated in someone’s honor without there being any benefit or return, and the sense of dapaticus, dapatice evokes the idea of profusion, of what one “sacrifices” to make a display of generosity in the treatment of a guest. The Latin daps and the Greek dapánē thus have in common the feature of a lavish expenditure on the occasion of a religious feast, of a “sacrifice.” The notion of “expenditure” is by no means a simple one (cf. above on the “gift,” in Book One, Chapter Five).
Given the clear connection of form between dapánē and damnum, it remains to see how the connection of sense can be explained. Damnum is primarily “expenditure,” as emerges clearly from Plautus (Miles 699): a character complains of financial embarrassment brought on him by marriage, of the expenses occasioned by his wife, haec atque eius modi damna: these “expenses” which are really a “loss of money,” a damnum. This sense persists in the adjective damnosus, which means nothing more than “extravagant”; and finally in damnare itself, again in Plautus. Here is one example of many (Trinummus, 829, a prayer to Neptune): “Haven’t you heard it said that people say in your honor” pauperibus te par cere solitum ‘that you have of the custom of sparing the poor but’ divites damnare atque domare ‘you hit the rich in their pocket?’ Damnare here must be understood as “to compel to spend,” expenditure always being regarded as a “sacrifice” of money.
Here we have the origin of the sense of damnum as “damage”: it is properly money given without any return. Damnare does not primarily mean to condemn in general, but to compel someone to spend money for nothing.
Daps, which has a religious sense, like the words connected with it in Armenian and Icelandic, throws light on the meaning of the terms related to it and also receives some illumination in return: it means “sacrifice” but also “a ceremony on the occasion of a festival.” According to an ancient rite, after the conclusion of a ceremony, by way of pure ostentation, a meal was offered which involved a great deal of expense, which diminished the fortune of the person offering it but gave him the satisfaction of honoring his guests and being honored himself by his generosity.
In this way we can account for the relation between notions which became specialized either in law, like the Latin damnum, or in economic life, like the Gr. dapánē.
This review of the terms relating to sacrifice may also include the Greek thúō ‘sacrifice’, with the numerous derivatives made from it. Its origin is certain: thúō goes back to a present tense *dhu–yō the root of which properly means “to produce smoke,” and it is directly related to the Latin suf–fiō ‘to expose to smoke, to fumigate’. A confirmation of the etymology is brought by a Greek derivative, the relation of which to *dhu– is, however, not obvious: this is the word for “sulphur,” the Homeric théeion or theîon, which naturally has nothing to do with the adjective theîos ‘divine’, as is clearly shown by the Homeric form. It is derived from the root by means of the suffix –s and goes back to an ancient form *dhwes–ion, cf. the Lithuanian present stem dvesiu ‘breath, pant’.
The word for “sacrifice” in Greek thus goes back to the idea of “fumigation,” the fat which is burnt, the exhalation of the flesh which is roasted, the smoke which rises and ascends as an offering to the gods: a conception of which the Vedic and Homeric texts offer numerous examples.
If this etymology throws some light on the notion of “sacrifice” in Greek, it may also illuminate a family of Latin words which are probably related to it. Starting with a form with a suffix –ro, *dhwes–ro, we get in Latin the stem febro-, februum and februare, together with the noun februarius. The whole group relates to “purification,” a function which is illustrated by specific rites: februarius, the month of purifications, is the last month of the old Roman year. This “purification” is etymologically a “fumigation,” the intermediary being the Greek term for “sulphur,” for sulphur was used to purify by fumigation.
The prehistory of these two important lexical groups may thus be illuminated by a comparison which strives after the highest degree of rigor. Nevertheless it must be insisted that certainty has not been reached. For the derivation of febro-, for instance, a Latin f- may have a number of origins, and the internal –br– could also be interpreted differently. Hence it cannot be proved that febro– may not have a different origin than *dhwes–ro-. It is sufficient that this provides a probable explanation.
If we examine the terms which denote “purification” in Latin, we may single out another because it raises a problem which has been much discussed: this is lustrum, lustrare. This was the term given to a ceremony which every five years served to purify the people assembled on the Campus Martius and gave rise to solemn rites accompanied by a military review. Under lustrum we distinguish three lexical units: lustrum, a period of time, the five-year interval between successive performances of this ceremony; lustrare ‘to review’ (e.g. perlustrare oculis ‘to survey an object’, ‘to allow one’s eyes to rove over’); and lustratio, ‘purification’.
There has been much discussion of the proper meaning, the etymological meaning, of these words. Two explanations have been advanced which we must briefly discuss. One suggests that lustrum has a connection with the root that means “to shine,” that of lux, which produces the verb illustrare with the adjective illustris, which is probably a derivative of it. Now lustrare and illustrare cannot be dissociated, in point of form, nor associated in point of meaning. Illustrare can be explained directly from lux but does not show any of the technical senses of lustrare. Similarly the neuter lustrum could go back to *loukstrom, just as luna does to *louksna. But since for semantic reasons there appears to be no connection between illustrare and lustrare, efforts have been made to find a different explanation for lustrum. The proposal has been made to connect it with the root which means “to wash,” loúō in Greek. But lustrum shows no trace of the proper sense of loúō ‘to wash’: to wash is not to purify, and the lustrum is not characterized by the kind of purification which is brought about by the use of water either in the form of aspersion or immersion. There is also a phonetic difficulty. If we trace the word back to the root of loúō we should posit an ancient *lowestrom and this would give *lōstrum as a regular development. In that case we should have to regard lustrum as a dialect form.
In default of a definitive explanation we may try to delimit the exact sense of the term.
The most explicit text is very short (Livy I, 44). It relates to the foundation of the ceremony of the lustrum, at the time of the first operation of the census. The rite is said to have been instituted on the occasion of the census proclaimed by Servius Tullius. After the census had been taken, Tullius commanded all the citizens to present themselves on the Campus Martius drawn up in their centuries:
Ibi instructum exercitum omnem suovetaurilibus lustravit, idque conditum lustrum appellatum, quia is censendo finis /actus est. ‘Once all the troops had been lined up, he purified them by the suovetaurilia; and that was called the conditum lustrum because it was the end of the taking of the census.’ Conditum lustrum is translated as the “conclusion of the lustrum.” But the preceding sentence contains an indication which ought not to be neglected: “edixit ut omnes cives Romani … in Campo Martio prima luce adessent.” The citizens had to present themselves at dawn, on the Campus Martius, formed up in centuries, both infantry and cavalry. It is, therefore, probable that prima luce was a ritual condition of the ceremony and not a fortuitous circumstance.We know how the lustratio was performed. The purifiers, priests or kings, made a circuit round the group of people or the building which was to be purified, always proceeding towards the right. Thus the purification occasioned a circumambulation: consequently lustrare denoted “to traverse, to review” as well as “to purify.” If we could connect lustrare with the prima luce of the preceding sentence, an explanation would emerge: lustrare would be literally “to illuminate.” The procession would then be the imitation of the sun which with its rays illuminates in a circular way. There would be a correspondence between the circumambulation of the priest and the circular motion of the star.
Such an explanation, which is the simplest from the etymological point of view, would be founded on the facts and would agree most simply with the tradition. Once the circumambulation was finished and all the people reviewed, the census was taken: is censendo finis /actus est.