Use the following persistent identifier: http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Benveniste.Indo-European_Language_and_Society.1973.
Chapter 6. The Censor and Auctoritas
If the Roman magistrate with specifically normative functions is called censor and if the senators whom he enrolls formally register their authoritative opinion by saying “censeo,” this is because the IE *kens– strictly meant “to affirm a truth (which becomes law) with authority.”
This authority—auctoritas—with which a man must be invested for his utterances to have the force of law is not, as is often stated, the power of promoting growth (augere), but the force (Skt. ojaḥ), divine in principle (cf. augur), of “causing to exist.”
We have established a frequent relation between terms used with reference to institutions and verbs which denote in one way or another the idea of “to say.” There is often a close connection between the act of speech and law or rule insofar as they serve to organize certain social functions. In particular, political institutions are sometimes called by terms which involve some specialization of the notion of “speech” in the direction of authoritative pronouncement. Thus the diversity of the notion of “speech” is illuminated by a study of the words used with reference to it. We see that the terminology of speech proceeds from a variety of origins and concerns very different semantic spheres. The work of the comparatist can be instructive in determining the point of departure for the terms which denote “to say” that have become words denoting institutions and names for authority.
We may take a new example, peculiar to Italo-Celtic and Indo-Iranian, one of those words which throw light on inter-dialectal relationships and attest survivals of a cultural nature: the Latin censeo, censor, census.
The censor is a magistrate, but the verb censeo means no more than “estimate, judge, pronounce an opinion”; whereas census is a technical operation, the assessment of the wealth and classification of the citizens. The same verb is known elsewhere than in Latin, in one of the Italic languages. In Oscan we have the infinitive censaum ‘censere’ and also a noun kenzstur, kenzsur ‘censor’, probably imitated from Latin. On the other hand the corresponding stem in Indo-Iranian has given rise to a considerable number of verbal and nominal forms with a marked difference of meaning. This is the root seen in Skt. śams– ‘praise, pronounce a eulogy of’ and of the abstract śasti ‘praise, eulogy, recitation of hymns’. Parallel with Sanskrit śams– we have in Iranian: (1) Avestan saŋh– ‘proclaim solemnly, pronounce’, (2) Old Persian θanh– and θah-, which is usually translated as “to proclaim.” On this basis we can posit an Indo-European verbal stem *kens– the sense of which, according to all the dictionaries, was “proclaim solemnly.”
However, the very precise sense of the Latin terms can hardly be reconciled with so vague a definition, which incidentally would also suit equally well a number of other verbs. The magistrate called the censor had as his primary function the duty of making a roll of the citizens. It was the census, the assessment, which gave the term censor its whole meaning. To evaluate the private fortune of each person and assign each to his appropriate class is a hierarchizing function which must be derived from a root with an already specialized sense.
The censor was also concerned with the recruitment of the senate (lectio senatus). He also had the task of supervising the morals of the citizens and repressing excess of every kind: the breaking of moral rules and the correction of excessive luxury and extravagance. It was from this that censura got its moral sense. Finally the censor was charged with placing the contracts for farming the taxes, with public works, and with regulating the relations between the contractors and the state. All these different functions are in some way connected with the essential function of the censor, which was the census, the classification of the citizens.
The verb censeo is used in a formula which is often quoted (Livy 1, 32, 11-12). In the procedure for the declaration of war established by Numa, the rex consulted each of the Fathers of the Senate: dic quid censes—and the other would reply: puro pioque duello quaerendas (with res understood) censeo. ‘I am of the opinion that we should seek what is our due by a just and holy war’. By this formula the Father in question pronounced in favor of the war and underlined its necessity. This same verb also figures in the rule laid down by the senatus consulta ‘the decrees of the senate’.
In describing these uses we could content ourselves with translating censeo by “judge, think, estimate.” But the nouns from the same root, censor and census, require a more precise sense which must reflect the real sense of the Indo-European root. G. Dumézil  has applied himself to the task of giving a precise sense to the root. He has sociologized the notion of śams– in a definition, which is also valid for Indo-European and which already contains in essence the Roman census:
The technical sense of censor and census must not be a secondary sense but must preserve what is essential in the primary meaning. At the outset we must doubtless posit a politico-religious concept such as: to site (a man or an act or an opinion, etc.) in its correct place in the hierarchy, with all the practical consequences of this situation, and to do so by a just public assessment, by a solemn act of praising or blaming (p. 188).
Unlike the usual translation, we have here a definition of great precision, the result of which is to take back to the Indo-European common period the sense of the Latin census, censor. It seems to us that this definition, if we posit it as Indo-European, includes some elements which owe their inclusion to perhaps too close a reliance on the sense of the Latin words.
A study of other words of the same root, particularly in Iranian, leads us to a rather different view, which takes more account of the different senses which are attested. It will be useful to analyze the evidence offered by Old Persian.
(1) In the inscriptions the king uses the verb corresponding to the Skt. śams-, Lat. cens– in the form of the third person of the present stem θātiy to introduce his own speech. He introduces each section of the text by the formula: θātiy dārayavahuš xšāyaθiya ‘thus speaks (proclaims, pronounces) Darius the King’. There follows a text of variable length and then the formula recurs to introduce a new topic, and so on until the end of the next. This is the set way of composition in use during the whole of the Achaemenid period.
(2) Darius enumerates his ancestors back to the eponymous Haxāmaniš (Achaemenes) and says: “this is why we call ourselves (θahyāmahiy) Achaemenids.”
(3) Darius boasts of the submission of the peoples who have remained faithful to him and of the solidity of his power: “Everything which was commanded to them and prescribed (aθahiya) by me, this they carried out whether by day or by night.”
(4) Darius comes to the subject of the Magus Gaumāta, the false Smerdis of Herodotus. This magus falsely usurped the kingship by deceiving his subjects. He was greatly feared because of the massacres which he had ordered “and no one dared to say (θastanaiy) anything against him.”
(5) Then comes the list of all the rebels who have usurped royal authority. Each one is evoked in the same terms “such and such rebelled; he usurped power saying (aθaha): I am so and so, the sole legitimate king.”
(6) At the end of the inscription, after telling of his accession to the throne and expounding his political acts, Darius addresses the future reader: “If you read this inscription and you get others to read it and you tell (θāhy) what it contains, Ahura Mazda will protect you and your lineage will be long. If you conceal the contents of this inscription, Ahura Mazda will strike you and you will have no descendants.”
(7) Finally in an inscription called the “Testament of Darius,” the king proclaims the rule which he will follow with regard to what a man says (θātiy) against another man.
We have now gone through all the forms and uses of the verb. Certainly, on a cursory reading, we could make do with equivalents, according to the passage in question, such as “say, proclaim, prescribe” and elsewhere, “call oneself.” But we should try and give a closer definition of the sense. The most frequent use (1) is not the most instructive. Light will be thrown on this formula by other uses. Let us take rather (4): No one dared to “say” anything against Gaumāta, because they feared him. There is another verb for “to say” in Old Persian (gaub-). In the above passage what is meant is that no one dared “to tell the truth” (many people were aware of the identity of the usurper and Gaumāta had put numerous persons to death for fear of being recognized); thus “to say” in this connection is analytically “to say who he was in reality.” Similarly, with (5): the rebel chieftains falsely assumed the title of king. They “spoke” (untruthfully); however, they claimed to be telling the truth, and their assertion was an emanation of authority.
Next we have (6): if you make this proclamation known to the people, if you “say (what it contains),” that is, if you report its true content.
(7) concerns what a man “says” against another; such an utterance claims to be true, and it may entail legal consequences.
We now return to usage (2): after having enumerated his ancestors back to the eponymous Haxāmaniš (Achaemenes), Darius concludes: “this is why we call ourselves Achaemenids”; this is a statement of dynastic legitimacy; we proclaim the fact of being Achaemenids as our true and authentic status.
We now come to the last, the most usual use of the verb, the use to introduce each section of the text. The king θātiy; he “proclaims” what is the case: Darius maintains what is true, both in the reality of the facts which he relates and in the reality of duties towards Ahura Mazda and towards the king. It is both a factual and a normative truth.
Thus at the conclusion of this review of the evidence we reach a definition of the verb which we might put thus: “to assert with authority as being the truth; to say what corresponds to the nature of things; to proclaim the norm of behavior.” He who “speaks” is thus in a position of supreme authority; by declaring what is, he fixes it; he proclaims solemnly what is imposed, the “truth of fact or duty.”
Such is the witness of one of the Indo-European languages, Old Iranian. The evidence of Old Persian is confirmed by the uses of saŋh– in the Avesta, whereas in Vedic the semantic development is directed towards religious proclamation: śams– ‘proclaim, praise’.
We may now return to censeo. Our definition makes intelligible the specialization of sense undergone by censeo, census, censor in Roman institutions. In that he establishes with authority a factual truth, the censor proclaims the situation of each citizen and his rank in society. This is what is called the census, the assessment which establishes a hierarchy of status and wealth. More generally, censeo means “to assess” everything according to its true value, hence both “to appraise” and “to appreciate.” To perform this function the requisite authority is needed: hence the question quid censes? which was ritually put to the senators by the king.
Censor has a complementary notion which is constantly associated with it in the uses of the word in Latin and which is implied by our definition, that of “authority”: censeo is often collocated with auctor and auctoritas.
What is the significance of these words and what is its etymological foundation? It is clear that auctor is the agent noun from augeo, which is usually translated “grow, increase.” To augeo corresponds the Greek present tense auxánō and, on the other hand, the alternative form *weg-, represented in German wachsen and English wax (opposite of wane). In the guise of these two alternating forms of the root the Indo-European stem means “to augment, grow, increase.” But the Indo-Iranian correspondents are exclusively nominal: Skt. ojaḥ, a neuter in –s, ‘might, power’, in Avestan aogar-, aoǰah– ‘might’, and the adjective Skt. ugra-, Av. ugra– ‘strong’.
In Latin, besides auctor, we have an old neuter which has become masculine in the shape of augur, with its derivative augustus; these words form a group apart.
We now see the double importance of this group of words. They belong to the spheres of politics and religion and they fall into a number of sub-groups: that of augeo, that of auctor, and that of augur. It would be of interest to find out how the notion of “authority” came to be derived from a root which simply means “grow, increase.” Our dictionaries, which translate the verb with this meaning, also define auctor as “he who causes to grow, the author.”
This definition may appear strange and in any case it is inadequate. We are invited to believe that the profound meaning of auctor could be simply traced back to the notion of “growth.” This is hardly satisfactory. The notion expressed by auctor and its abstract auctoritas is difficult to reconcile with the sense “to increase,” which of course is indubitably that of the verb augeo. But is this the primary sense of the verb augere? We leave augur for the moment, to come back to it later on. The fact that in Indo-Iranian the root aug– means “might” is noteworthy. Further, Skt. ojas-, like Av. aoǰah– and their derivatives, refers in particular to the “might” of the gods; the Avestan adjective aoǰahvant– ‘endowed with might’ is almost exclusively used of gods. This implies a power of a particular nature and effectiveness, an attribute which belongs to the gods. But we disregard the sense peculiar to Indo-Iranian and confine ourselves to Latin. The problem here, as it so often is, is to give an exact definition of the real sense of the basic term, in such a way that the derivatives find herein their own explanation. Now the sense of auctor in its different uses cannot be derived from that of “increasing” which is assigned to augeo. A large proportion of the senses of augeo remain in the dark, and this is precisely the essential part, that from which the special applications have developed so that they have in the last resort ended up by splitting off into distinct units.
Scholars persist in translating augeo as “increase.” This is accurate for the classical period but not for the earliest texts. For us “to augment” is equivalent to “increase, make something which existed before bigger.” Herein lies the unnoticed difference from augeo. In its oldest uses augeo denotes not the increase in something which already exists but the act of producing from within itself; a creative act which causes something to arise from a nutrient medium and which is the privilege of the gods or the great natural forces, but not of men. Lucretius often makes use of this verb when he is retracing the genesis of beings in the universal rhythm of birth and death: quodcumque alias ex se res auget alitque ‘whatever thing gives rise to other things from itself and nurtures them’ (V 322); morigera ad fruges augendas atque animantis ‘prone to engender plants and living creatures’ (V 80). In the archaic prayer formulas the Romans also used augere of the benefits they expected from the gods, namely of “promoting” all their enterprises: Divi divaeque…, vos precor quaesoque uti quae in meo imperio gesta sunt, geruntur, postque gerentur, … ea vos omnia bene iuvetis, bonis auctibus auxitis ‘Ye gods and goddesses, I pray and beseech you, that whatever has been done, is being done, and shall be done hereafter under my imperium, you shall aid all those things and increase them with good increases’, that is, cause them to prosper (Livy 29, 27).
Much the same sense is evident in the uses of the agent noun auctor. The term auctor is applied to the person who in all walks of life “promotes,” takes an initiative, who is the first to start some activity, who founds, who guarantees, and finally who is the “author.” The notion expressed by auctor is diversified according to the different contexts in which it is used, but they all go clearly back to the primary sense “cause to appear, promote.” This is how the abstract auctoritas acquired its full force: it is the act of production or the quality with which a high magistrate is endowed, or the validity of a testimony or the power of initiative, etc., each of these special applications being connected with one of the semantic functions of auctor.
The religious term augur may also be linked to augeo. This was already the opinion of the Latins. Augur would have been an ancient neuter which designated at first the “promotion” granted by the gods to an undertaking and made manifest in an omen. This confirms that the action of augere is of divine origin. From *augus, a doublet of augur, is derived the adjective augustus, literally “provided with *augus,” that is to say, “endowed with such (growth)-promoting power.”
From an early date this single semantic unit broke up into five independent groups: (1) augeo with augmen, augmentum, auctus; (2) auctor with auctoritas, auctoro; (3) augur with augurium, auguro; (4) augustus, a title which became a proper name and then produced augustalis, augusteum, etc.; (5) auxilium with auxilior, auxiliaris.
The primary sense of augeo is discovered in auctoritas with the help of the basic term auctor. Every word pronounced with authority determines a change in the world; it creates something. This mysterious quality is what augeo expresses, the power which causes plants to grow and brings a law into existence. That one is the auctor who promotes, and he alone is endowed with the quality which in Indic is called ojaḥ.
We can now see that “to increase” is a secondary and weakened sense of augeo. Obscure and potent values reside in this auctoritas, this gift which is reserved to a handful of men who can cause something to come into being and can literally ‘bring into existence’.
[ back ] 1. In his book Servius et la Fortune, “Essai sur la fonction sociale de Lounge et de Blâme et sur les éléments indo-européens du cens romain,” Paris, 1943.