Use the following persistent identifier: http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Benveniste.Indo-European_Language_and_Society.1973.
Chapter 7. The Quaestor and the *Prex
Lat. quaero ‘seek, ask’ (whence quaestor, quaestus), a word without an etymology, has close connections with precor, *prex ‘to pray, prayer’ which must be pinned down: in fact it is not only in Latin that the two terms seem to form a redundant combination, as in the old formula “Mars pater, te precor quaesoque,” but in other languages too, derivatives from *prek– (Iran. frasa, OHG forscōn) have exactly the same sense as the Latin quaero. In the absence of decisive pointers in the languages in which only *prek– is represented, it is only in Latin that a distinction appears: as opposed to *prek-, which denotes a verbal request (precor, procus), the group of quaero, quaestus ‘means of gaining, gain’, quaestio, ‘question, torture’, quaestor ‘examining magistrate’ and ‘tax collector’ is defined by the material and non-verbal character of the methods used to obtain what is being sought.
In the terms studied up till now it has been etymology which helped us to determine the primary sense which is the source of the others. But there are instances where etymology fails us; in such cases our sole recourse is to traditional stock uses. It is in such conditions that oppositions of vocabulary can operate, those differentiations which, by establishing a connection between two terms, enable us to distinguish and illuminate the terms involved.
Now in the lexical series under examination, in the Latin vocabulary in particular, two words present themselves: one is the verb quaero, the other is the agent noun from this verb quaestor. The sense of the verb is a general one, whereas that of the noun is specialized. Quaero is translated as “to seek,” the quaestor is a magistrate who had the dual function of “examining magistrate” and “warden of the treasury.” In the judicial language quaero meant “to make an enquiry, investigation,” and in this sense it was the equivalent of the Greek zēteîn. However the accepted meaning of the verb does not account for the sense which the noun quaestor has.
Further there is a verb which in other languages conveys the same sense as the Latin quaero: this is the verb whose root appears in the Latin precor, *prex. In Latin there is a difference between the verbs quaero and precor, but elsewhere we find forms of the root corresponding to precor to designate the kind of activity in which the quaestor specialized. Here we have one of those problems when two verbs of similar sense have been specialized in different ways in different languages. Only the conditions of their use can enlighten us in the absence of any etymology.
Let us first consider quaero by itself and in its relation to quaestor. The quaestor was properly the magistrate whose full title was quaestor paricidi et aerari. The function of the quaestor as the guardian of the finances of the state (aerarium) was secondary to his first function, cf. Festus (247, 19): parricidi quaestores appellabantur qui solebant creari causa rerum capitalium quaerendarum ‘The name parricidi quaestores was given to those who were appointed to investigate capital offences’.
It will be noticed that quaero is expressly used in the formula which explains the noun quaestor. Here we already see a technical use which invites us to interpret with greater precision the sense of the verb: we have to start with a special use of quaero in order to find the sense of quaestor, especially in the title quaestor paricidi.
Here we shall have to make a digression on the subject of paricidium and paricida. In the last few years a series of different interpretations have been put forward in explanation of this very ancient word on which the Romans themselves have no very clear opinion. In the first place we have the etymology which identifies the first component with pater. This is certainly to be rejected. Today a number of comparatists would see in the first term of paricida a word signifying “man” in general. This is the thesis of Wackernagel  who starts from the idea that paricida is a general term for the murderer of a man. Pari-, according to him, is a word for “man,” unknown elsewhere in the western vocabulary, but corresponding to Skt. puruṣa ‘man’. There is no great formal difficulty about this equation, if we admit that puruṣa goes back to *purṣa. But what constitutes an obstacle to the acceptance of this equation is the sense of the Latin compound and its use in the legislation of the Romans.
In our view we should retain the old etymology which equates pāri– with the Greek pēós (originally *pāso-): it has been taken up again and justified on a number of occasions, most recently by L. Gernet,  who by means of juridical arguments shows that we must adhere to this interpretation.
The Greek term pēós properly designates “the kinsman by alliance, by marriage.” Thus in the Iliad (3, 163) we see it associated with phílos, which has the full sense studied above.  In the Odyssey (8, 581ff.) we find it used with other kinship terms which explain it: “Have you a pēós who died before Troy, a son-in-law or a father-in-law, those who are dearest to us after those of our own blood and our own race? Or was he a dear companion? For it is better to have a companion full of wisdom than a brother …” Thus pēós is, linked on the one hand with gambrós ‘brother-in-law’ and pentherós ‘father-in-law’, and on the other with hetaîros ‘companion’ or phílos: it is therefore someone with whom one has contracted an alliance. This is the category of kinship which pēós defines: it is kinship by alliance, within a tribe. This kinship imposes certain precise obligations, notably in the case of violence done to one of the parties concerned.
We may now examine the famous text of Numa Pompilius on the parricide (Festus, loc. cit.): “Si quis hominem liberum dolo sciens morti duit, parricidas esto.” In this text, as in all the codes and rituals at Rome, the words must be taken in their full sense. The man who puts to death with malice aforethought a man of free birth must be a parricida, must be considered as “the murderer of a kinsman by alliance.”
There are, as we have seen, certain legal provisions which concern simply the family, and there is on the other hand inter-family law which regulates the relations between different families. One might say that thémis and díkē are involved in the semantic context of this provision. We see that one who kills a liber man is treated as a parricidas; the notion of a murderer within a family is extended to the case of a murderer within society itself. Homicide in general is not punished as such in the ancient law codes. In order to be punishable it was necessary for the murder to affect a man of the group: morality stopped at the frontier of the natural group.
Thus the quaestor paricidi exercised his functions within the social group which was considered as being the family group in its full extension including its connections by alliance. With the help of this closer definition of the sense we can now attempt to give precision to the meaning of quaero. The meaning “to make an investigation” is evidently too closely linked with quaestor and its derivatives to be posited as the primary sense. It will be better to start from an example which has every mark of antiquity and authenticity.
This is an old prayer (Cato Agr. 141), an invocation to Mars pater on the occasion of the lustration of the fields. This text, which is important in itself, is full of archaisms and has been preserved to us in its original state.
In it we find a reference to the sacrifice called “su–ove–taurilia,” a term which has been analyzed above  and which reveals a profound social symbolism. Neither the order nor the nature of the animals is accidental. We have here three symbolic animals: the pig is sacred to the divinities of the earth, to Ceres: it is associated with the fertility of the soil; the bull is traditionally sacred to Jupiter and to Zeus; it is the animal offered in the most sacred and solemn sacrifices, those which are in the charge of the priests of the highest divinities. Coining between these two we find often if not always, the sheep, the ram which is the animal of the warriors. We have here precisely the three social classes, represented by symbolic animals. This is what gives the key to the sacrifice of the lustration. The sacrifice called “suovetaurilia” united symbolically the three orders of society in this solemn communion for the protection of the great god who is invoked, Mars: and the society as a whole which makes the sacrifice is represented at the ceremony.
This symbolism reveals the archaism of a prayer like this. It begins with this invocation: Mars pater, te precor quaesoque uti sies volens propitius … “I beg and beseech you”: is this a simple duplication? Some scholars have reproached the religious language with redundancy: the terms look as if they were duplicated and even triplicated, as though the authors had the purpose of accumulating equivalents. But this is not the case. On closer examination we see that these juxtapositions do not in fact associate terms of identical, or even closely related sense; each one keeps its full sense and this is a condition for the effectiveness of the prayer.
A second example is provided by Lucretius: prece quaesit (V 1229) ‘he asked with prayer’. Such examples in which *prex and quaero are collocated are most instructive for our analysis.
Finally, and this is especially important, we must ask how the verb quaero and its frequentative form quaeso ‘ask persistently’ are employed. We have had occasion to examine from a different point of view the formula which in ancient Roman law summed up the purpose of marriage: liberum(-orum) quaesundum(-orum) causa (gratia) ‘to obtain (legitimate) children’;  we can hardly translate the verb otherwise than as “obtain.” In any case the sense is certainly not “ask insistently, to pray repeatedly.”
Finally the nominal derivative quaestus, in its usual application, denotes “gain, profit” and also “the way one earns one’s living, profession.” This term falls completely outside the legal series which begins with quaestor and continues with quaestio ‘(judicial) investigation’ and also “torture” (whence quaestiono ‘investigate by means of torture, to torture’). Here then is the list of the principal terms of the semantic group of quaero, with the variety of meanings which they present.
To achieve further precision we must now turn to the verb with which it is associated: precor. This present tense is derived from a well-known root *perk-/*prek-, which is widely represented in both stem forms without difference of meaning. In Latin we have *prex, precor, posco (the inchoative present of preco), postulo. The Romans remained conscious of the connections between these forms as well as of the difference of sense which each one specifies.
Outside Latin we have (1) the verbal stem Skt. pr̥ccha– ‘ask’, Iranian pr̥s– (< *perk-) and fras– (< *prek-); OSl. prositi, Lith. prašýti and (2) a noun Skt. prāt– (vivāka) ‘judge’, literally “he who decides a prāt”. The sense is restricted in an instructive way, for prāt is a “question” in the legal sense; it is the “case,” that is, the semantic equivalent of the quaestio of the quaestor. To Skt. prāt corresponds also the OHG frāga ‘question (Germ. Frage)’, a term which differs from *prex only in the root vowel ā.
(3) In a different semantic sphere we have Lat. procus, ‘he who demands’ in marriage, the suitor. This specialization of sense recurs in the Lithuanian pir̃šti ‘ask in marriage’.
(4) With the present morpheme –ske-, known from Latin posco, we have the Avestan and Persian frasa ‘make an investigation, ask’ and also “punish, chastise”: avam hufraštam aprsam (where hufraštam contains the participle frašta– of the same verb) ‘(he who has disobeyed me, says Darius), him I questioned (in such a way that he was) well questioned’, which means “I punished him severely.” Finally we have OHG forscōn ‘seek, make an investigation’ in speaking of a judge.
We see, then, that in a number of languages particular forms and uses of *prex– coincide with those of quaero, but always outside Latin: in Sanskrit, Iranian, and Old High German.
|Latin *prex||Skt. prāt-vivāka||Lat. procus|
|OHG frāga||Lith. pir̃šti|
|(cf. Skt. pr̥s-, Iran. fras)|
In Latin itself, however, as we have seen, the two verbs are associated in such a way that their meanings seem to be closely akin. We can see how far they coincide and how they differ. In two cases the context was the formulation of a request, but in two different ways: precor, *prex must be taken together with the agent noun procus ‘he who asks in marriage’; *prex is the request which is exclusively verbal and is especially addressed to gods to obtain what one hopes for from them. Such is the distinctive character of prek-: it is an oral request, addressed to a superior authority and which does not comprise any other means than speech.
On the other hand, quaero, with the derived nouns quaestio and especially quaestus, denotes a different procedure: quaestus ‘way of making a profit, profit’, quaestio ‘interrogatory torture’, and the verb quaero itself involves not the attempt to get information or other things by oral request but to obtain something by the appropriate material means.
It is not precisely some information that is solicited or a favor that is requested but rather some material object, often some advantage, but always something concrete, which is considered necessary to life or activity.
This is confirmed by an expression like liberum quaesundum causa: seek to obtain (and not to know). The quaestus and the quaestio show this no less clearly and it is also apparent in quaerere victum ‘get one’s living, earn one’s living’, and quaerere rem ‘to get rich’. We also read in Terence: hunc abduce, vinci, quaere rem (Ad. 482): ‘take him away, bind him and get something from him’, that is to say, “extract the truth from him by appropriate means.” They seek to gain by some material means something which is vaguely referred to as res. What is relevant here is only the means employed for obtaining it; it is not simply a matter of asking.
Thus the formula precor quaesoque is by no means a tautology or a rhetorical duplication. Precor is to ask by means of *prex. Here speech is the intermediary between the one who asks and the one who is asked. Speech is by itself the effective means. But quaeso differs from precor in that it implies the use of means appropriate to getting what is desired, like the sacrifice of the three animals and the association of the formula with the offerings.
To achieve this reconstruction we have had to use the forms of *prek– which occur in languages other than Latin, especially Iranian. We have stressed above that Iran. fras, frašta take on the sense of “punishment” and generally “torture.”
We can now return to our point of departure, which was the Latin title quaestor. It is now clear that the quaestor was not merely charged with “making an enquiry”; his role was rather quaerere, to try and obtain by material means, either, in a criminal case, the person of the guilty party, or (and the word is associated with quaestus) money for the treasury, for the incomings and outgoings of which he was responsible.
Such is the meaning which we propose (based on the uses of the verb) for the agent noun quaestor. In the example from Lucretius, prece quaesit, there is also no tautology: the object of quaerere is pacem, and this is the material object which he seeks to gain: by what means? By *prex, by an oral request. In other circumstances other means would have been employed.
Thus we have established a duality of function which betrays an ancient functioning. For us “to request” is “to seek to obtain.” This notion is specified in different ways according to the context. But in Old Latin two different notions of asking were distinguished: in ancient societies they had a precise and concrete form and only the vocabulary can reveal this to us. The verbs or certain of their derivatives preserve for us, or yield by the application of the comparative method, the evidence of more complex semantic distinctions: such is the gap between procus and precor because of their early specialization. If we did not know the senses which justify us in bringing procus into relationship with Lith. pir̃šti, it would be difficult for us to give the root *prek– its exact sense, and to see that *prek– denotes a purely verbal activity, not employing any material means and consisting of a request generally addressed by an inferior to a superior. It is thus that *prek– ‘ask a favor’ differs from the root, not attested elsewhere, which is represented by the Latin verb quaero and the agent noun quaestor.
[ back ] 1. Gnomon VI, 1930, p. 449 ff. (= Kleine Schriften II, 1302 ff.)
[ back ] 2. Revue de Philologie 63, 1937, pp. 13-29.
[ back ] 3. Book Three, Chapter Four.
[ back ] 4. Book One, Chapter Two.
[ back ] 5. Book Three, Chapter Three.