Giving and Taking in Indo-European Society

By Dr. Émile Benveniste (1902-1976)
Former Professor of Linguistics
Collège de France

Gift and Exchange


Greek has five words that are commonly translated uniformly by “gift.” A careful examination of their use shows that they do in fact correspond to as many different ways of envisaging a gift—from the purely verbal notion of “giving” to “a contractual prestation imposed by the terms of a pact, an alliance, a friendship, or a ‘guest-host’ relationship.”

The Gothic term gild and its derivatives take us back to a very ancient Germanic tradition in which the religious aspect (“the sacrifice”), the economic aspect (a mercantile association), and the juridical aspect (the atonement for a crime) are closely interwoven.

The varied history of the words related to Gr. dáptō, Lat. daps, on the one hand discloses the practice of “potlatch” in the Indo-European past, and on the other hand shows how the ancient notion of “prestigious expenditure” became attenuated to mean “expense, damages.”

The Hansa, which had become in the form of the guild an economic association, continues the tradition of the comitatus of young warriors attached to a chieftain, such as Tacitus described in his Germania.


We now approach the study of a complex of economic notions that is difficult to define otherwise than by the sum of their peculiarities: “gift,” “exchange,” “trade.” The terminology relating to exchange and gift constitutes a very rich chapter of the Indo-European vocabulary.

We begin with the notion of “giving.” One might think that this is a simple idea. In fact it comprises some strange variations in the Indo-European languages, and the contrasts found between one language and another merit examination. Furthermore, it is extended to notions which one might not think of associating with it. The activity of exchange, of trade, is characterized in a specific way in relation to an idea which appears to us different, that of the disinterested gift. In this light exchange appears as a round of gifts rather than a genuine commercial operation. The relationship of exchange to purchase and sale emerges from a study of the terms employed for these different processes.

In this field there is great lexical stability. The same terms remain in use for very long periods, and, in contrast to what happens with those for more complex notions, they are often not replaced.

The Vocabulary of “Giving” in Greek

Little horse on wheels (Ancient Greek child’s Toy). From tomb dating 950-900 BC. Kerameikos Archaeological Museum in Athens

We start from the root *-, for which the consensus of languages guaranteed a constant form and meaning. The nominal forms show an ancient structure, that of derivatives in –no– and –ro-: Skr. dānam, Lat. dōnum, Gr. dō̂ron (δῶρον), Arm. tur, Slavic darŭ. These forms clearly attest, in the very constancy of this resemblance or of this difference, an ancient alternation r/n, this being the mark of an archaic declension, called heteroclitic, which is often revealed by the coexistence of derivatives in –r– and –n-. We have further a series of nominal forms in Greek, distinguished only by the class of derivation, all of which relate to “the gift.” They are: Gr. dṓs (δώς), dō̂ron (δῶρον), dōreá (δωρεά), dósis (δόσις), dōtínē (δωτίνη), five distinct terms which are uniformly translated as “gift.”

The first is very rare: we have only one example. The other four are much more common and may coexist in the same author. Is this a fortuitous lexical redundancy or are there reasons for this multiplicity? Such is the problem which we must investigate.

The first form, dṓs, is a stem in –t-. It corresponds to Latin dōs (stem dōt-). In Latin the word is specialized; it is the “dowry,” the gift which the woman brings into marriage, sometimes also the gift by the husband in purchase of his bride.

To establish the sense of Greek dṓs, which is not yet specialized, we have a verse by Hesiod: δὼς ἀγαθή, ἅρπαξ δὲ κακή, θανάτοιο δότειρα ‘dōs is good, but robbery (hárpaks) is bad, because it brings death’ (Works, 356). This verse is found precisely in a passage where the “gift” is highly praised as a means of establishing advantageous relations. Dṓs and hárpaks are root nouns, and it is no accident that no other example is found: they represent the idea in its most abstract form: “giving” is good, “robbing” is bad.

Dō̂ron and dōreá seem to have the same sense. But when Herodotus uses them concurrently we can see that he distinguishes them according to a principle which it is not difficult to discern. Thus III, 97: Κόλχοι…ταξάμενοι ἐς τὴν δωρεήν…δῶρα…ἀγίνεον ‘the Colchidians having assessed [having imposed payment on] themselves, brought gifts (dō̂ra) for the dōreá’. Dōreá is strictly the act of offering a dō̂ron. It is an abstract noun derived from dōréō (δωρέω), which is itself a denominative of dō̂ron. The verbal force is clearly seen in dōreá, and this explains the adverb dōreán (δωρεάν) (Attic) ‘by a gift, for a gift, gratuitously, for nothing’. Thus dō̂ron is the actual gift and dōreá is the act of bringing, of presenting a gift. From dō̂ron are derived dōreîsthai(δωρεῖσθαι) ‘to make a gift’ which governs the name of the thing or the person to whom it is given, and dṓrēma (δώρημα) ‘the thing which is presented, the present which serves as a recompense’.

Dósis is very different. Our translations do not distinguish it from dō̂ron; but its use in Homer Il. 10, 213 makes it clear: καί οἱ δόσις ἔσσεται ἐσθλή. A volunteer is needed for a dangerous mission; he is promised a good dósis, not a dō̂ron, because the object itself of the gift does not exist. Dósis is a nominal transposition of a verbal form in the present tense or, as here, in the future: “we shall give him, we shall make him a gift.” A formula where the verbal force of this abstract is still apparent is found in Homer, Od. 6, 208, δόσις δ’ὀλίγη τε φίλη τε —words used by people who give but who excuse themselves for not giving much. “This gift is a small one, but given gladly.” Thus dósis is “the act of giving.” The formation in –ti indicates an effective accomplishment of the idea, which may also, but not necessarily, be materialized in an object. Dósis may also designate a legal act. In Attic law it is the bestowal of a bequest, by express will, outside the rules of normal inheritance.

There is a further medical usage in which dósis denotes the act of giving, whence develops the sense of the amount of medicine given, a “dose.” Here the notion of gift or offering is absent. This sense passed by loan translation into German, where Gift, like Gr.-Lat. dosis, was used as a substitute for venēnum ‘poison’, whereas (Mit)gift ‘dowry’ still preserves the original connection with “giving.” In the early texts there is no interference of dósis either with dō̂ron or with dōreá.

Finally it remains to define the essential use of dōtínē. This is the most specialized term of the whole group. The examples are few but they are well characterized. It is a word of Ionic poetry which appears in Homer and also in Herodotus but soon passed out of use. Dōtínē certainly denoted some species of “gift,” but precisely what?

To persuade Achilles to return to battle, he is promised, among other things, a grant of land together with its rich inhabitants, who will become his subjects “…who will honor him (timḗsousi) like a god with dōtínai and who under his scepter will pay the liparàs thémistas” (Il. 9, 155–156).

The two words timḗsousi and thémistas are essential for defining dōtínēsi. By thémistas, an extremely complex notion, is understood the prerogatives of a chieftain; in particular, it is the respect shown, and the tribute brought, to a personage such as a king in accordance with the requirements of divine law. Still more important is the timḗ. This expression is derived from tíō and belongs to the group of Skt. cayati ‘to have regard, respect for’, from a root which must be strictly distinguished from that which signifies “to avenge, to punish,” Gr. poinḗ, which is often associated with it. Poinḗ, which corresponds exactly to Av. kaēnā‘vengeance, hate’ is the retribution that compensates for a murder. This also developed the emotional sense of “hate,” of vengeance considered as a retribution (cf. the sense in Iranian). The other group, the one which concerns us here, timḗ, denotes the honor due to a god, to a king, and the tribute due to them from a community. It is at the same time a mark of esteem and estimation in a social and sentimental, as well as an economic, sense. [1] The value attributed to somebody is measured by the offerings of which he is judged worthy, and these are the terms which elucidate dōtínē.

In Homer, Od. 9, 266–286, we find: “We are come to your knees to see whether you will offer us a xeinḗïon (a gift of hospitality) or whether you will give us a dōtínē, as is the law of hospitality (thémis xeínōn).” This time, in a text which seems made for our enlightenment, a relationship is established between dōtínē and the presents which are exchanged between host and guest according to the traditions of hospitality. Similarly, in Od. 11, 350 ff. “let our guest wait until tomorrow before leaving us so that I may be able to assemble the whole of his dōtínē .”

Fleeing from Athens, the followers of Pisistratus wanted to repossess themselves of the tyranny from which they had been ejected. To this end they travelled through all the towns which might have obligations towards them to assemble the dōtínai: ἤγειρον δωτίνας (Herodotus I, 61).

There exists also a verb dōtinázō, which is found once in Herodotus (II, 180): on the occasion of the reconstruction of a temple which was incumbent on a group of federated cities, the priests went from town to town collecting gifts: περὶ τὰς πόλις ἐδωτίναζον.

These quotations throw light on a very different notion from the others. It is not merely a present, a disinterested gift: it is a gift qua contractual prestation, imposed by the obligations of a pact, an alliance, a friendship, or a bond of hospitality; the obligation of the xeînos (of the guest), of the subjects towards their king or god and also the prestation implied by an alliance. Once this meaning has been established it helps us to solve the philological problems posed by the variants in the textual tradition of these words. Thus the manuscript tradition of Herodotus VI, 89 is divided between the reading dōtínēn and the reading dōreḗn. The Corinthians wanted to aid the Athenians and sold them twenty vessels, but at a very low price, at five drachmas per boat, because the law forbade a gratuitous gift. Thus it was a symbolic payment imposed on the Athenians because, according to the law, it was impossible for one city to give the vessels to another. Is this dōtínē or dōreḗ? In fact what was involved was a gratuitous present. The valid reading is therefore dōreḗn and not dōtínēn, which is excluded because it is a gratuitous gift which the law forbade, not that which is inherent in an alliance.

Such is the way in which the Greek distinguishes for the same notion “gift” between three nouns which, for all that they are derived from the same root, are never for one moment confused. This notion is diversified in accordance with social institutions and what I may call the context of intention: dōsis, dō̂ron, dōtínē, three words for expressing a gift, because there are three ways of conceiving it. [2]

A Germanic Institution: The Guild

German shoemakers’ guild, in Calle delle Botteghe in Venice

To the Greek terms we have reviewed we may now add the Germanic word which has become the name for “money,” in German, Geld. In Gothic, gild translates the Greek phōros ‘tax’, while the compound noun kaisara-gild translates the Gr. kē̂nsos‘tax’. We also have a verb: fra-gildanus-gildan ‘to render, repay, apodidónaiantapodidónai  and a derived noun gilstr, which likewise translates phóros ‘tax’.

In the other Germanic languages the sense is quite different: Old Icel. gjald ‘recompense, punishment, payment’; OE gield‘substitute, indemnity, sacrifice’; Old High German gelt ‘payment, sacrifice’; in composition gotekelt ‘Gottesdienst, divine service, worship’. In Frisian jeldejold appears the special sense which was to become generalized in Germanic: “a guild of merchants,” implying also “a corporation banquet.” The whole notion seems extremely complex within Germanic society; it is simultaneously of a religious, economic and legal character. We are here confronted with a very important question which dominates the whole of the economic history of the Middle Ages: the formation of the guilds, a problem so vast that it cannot be treated here and which in any case is more the concern of the historian than the linguist.

It is not the conception itself which we shall consider but rather the starting point of the great medieval economic associations which developed between the sixth to seventh century and the fourteenth century, especially in the coastal regions of the North Sea, in Frisia, in the south of England and in the Scandinavian countries.

The institutions have both an economic and a religious character. These fraternities were united by economic interests but apparently also by a common cult. They were studied in the work (1921) by Maurice Cahen, La libation en vieux scandinave. According to this scholar, toasts, banquets, compotations, were like rites which were celebrated by the members of a fraternity. This finally took on a specific form and became in Germanic countries an economic association.

The author, however, ran into a serious difficulty. According to modern historians of the Middle Ages, the guilds constituted an exclusively economic phenomenon of fairly recent date and did not reach back into the beginnings of the Germanic world. In these economic groupings, in which people were brought together by common interests, one should not look for a survival of older religious associations.

But more recent researches into medieval history have justified these conclusions. M. Coornaert has sketched the history of this institution in broad outline in two articles in the Revue Historique, 1948. Not content with confirming the ancient and religious character of the guild, he reproaches Maurice Cahen for deferring to the judgment of earlier historians who rejected any intrusion of comparative studies into this field.

At present the facts can be seen to form part of a continuous history which goes far back in time. It has been claimed that ghilda, a Latinized form of a Germanic word, does not go back further than the eighth century. But it is now known from the Gallo-Roman period in a text which is dated to 450 AD.

What is a guild? It is first and foremost a festive occasion, a sacrificial meal within a “fraternity” which has assembled for a voluntary communion, and those who are thus assembled bear the same name. The notion of a sacred banquet is at the very center of this expression. Now, we encounter it in 450, that is to say, shortly after the period when the Gothic text has become fixed in writing (towards 350).

It will thus be relevant to give close scrutiny to the Gothic data. The essential words gild and fra-gildan have no correspondents except in Germanic. It is a new term which offers no possibility of comparative study.

The Gothic word gild is found in the well-known question in Luke XX, 22: “Are we permitted to pay tribute to Caesar or not?” skuldu ist unsis kaisara gild giban…? In the same question Mark XII, 14 replaces gild with kaisara-gild. A neuter gilstr, that is to say, *geld-strum or *geld-trum, is given with the same sense: Epistle to the Romans, XIII, 6: “That is also why you pay taxes, φόρους τελεῖτε.”

The verb fra-gildan means “to give back, to restore,” Luke XIX, 8: “I give, gadailja (dίdōmi) to the poor (literally: I share my possessions with the poor); if anybody is cheated by me, I repay him fragilda (apodίdōmi ‘to make a return payment’ in the text) fourfold.” Cf. also Luke XIV, 12 and 14: “if you prepare a meal, do not invite your friends or your brothers or other relations, or rich neighbors, lest they invite you in their turn and this will result in an antapodόma, an obligation to further requital” (Gothic usguldan). The sense is “to render in exchange for what has been received,” not to give back the object itself but “to spend as much as is equivalent to the amount by which one has benefited.”

In order to understand the value of the terms in Gothic it is necessary to envisage the problem, which must have vexed the translator, of transposing into Gothic Greco-Roman ideas like those of Gr. phόros, Lat. census ‘tax, assessment, the obligation to obey a higher authority’, since the Germanic tradition knew only small independent groups, each obeying a particular chieftain, without any idea of a general organization.

Gild may be defined as a “reciprocal tribute”; it is a fee which is paid personally in order to benefit from a collective service within a fraternal grouping: an entrance fee (which is paid for in one way or another) into a fraternity bound by a common cult.

Wulfila thus gave a very special sense to a very different expression of traditional Gothic vocabulary, the word gild ‘obligatory contribution (paid to a group of which one is a member and a beneficiary)’, when he used it as an equivalent of phόros. This word evokes “a cult association,” a true fraternity which is fulfilled, expressed and reinforced in banquets and common celebrations at which affairs of high importance are decided.

In fact Tacitus (Germania, 22) speaks of the convivia of the Germans, the banquets which were an essential part of their social and private life. They attended under arms, a fact which simultaneously showed the military and civil character of the matters to be debated: it was there they discussed “the reconciliation of private enemies, the conclusion of family alliances, the choice of chieftains, peace and war, because they believed that there was no more favorable moment for man’s spirit to be open to frankness and to be fired to greatness.”

We have here the very important idea of the convivial communion, which is as it were the symbol and the intensification of the fraternity. The point of departure of the economic groups called ghildes lies in such fraternities which were bound together by a common interest, by one and the same activity. And within such a group the banquets, conviviaghilda are among the most characteristic institutions of the Germanic world. In thus “paying” (gildan) a fee to a fraternity, one pays a “due,” a sum which one must pay, and the payment itself is money, the geld.

We have here given a resumé of a long and complex history which has led up to institutions and to collective values. But to begin with the word was attached to an idea of a personal kind: the proof of this is the wergeld ‘the price of a man’ (with wer‘man’), the price which was paid for the expiation of a crime, the ransom. Let us take up once again the Germania of Tacitus, chapter 21: “they are obliged to share the hostilities of the father or their kinsmen as well as their friendships, but they are not prolonged indefinitely. Even homicide can be redeemed with heads of cattle which are a benefit to the household.” This wergeld, compensation for murder by a certain payment, is equivalent to Gr. tίsis; it is one of the ancient aspects of the geld.

We are thus on three lines of development: first religious, the sacrifice, a payment made to the divinity, secondly economic, the fraternity of merchants, and thirdly legal, a compensation, a payment imposed in consequence of a crime, in order to redeem oneself. At the same time it is a means of reconciliation. Once the crime is over and paid for, an alliance becomes established and we return to the notion of the guild.

It was first necessary to define these ideas in their mutual relations and their peculiar senses in Germanic in order to assess how far apart these terms were in their original meaning from the Greek terms which they were used to translate. This is a fact to which insufficient attention has been paid. Scholars have always tended to proceed by straight interpretation of the Gothic without noting the effort of transposition which must have been involved and the difficulties which resulted from it. These Gothic expressions, when compared with those in Greek, are quite differently structured.

Another difference lies in the manner in which economic ideas became established in the Germanic and classical languages respectively. They are often bound up with facts of religion which make still wider the gulf which separated them from each other in the past, and they took shape in wholly dissimilar institutions.

Prestigious Expenditure

Wealthy Ancient Romans entertained in lavish triclinia

We must remember that the fraternities constituted a group of close solidarity and a kind of dining club. The two aspects of this institution could be maintained also in other ways. What was in origin a convivial group might become with the evolution of society an association of an economic, utilitarian, and commercial character.

One of the two aspects, the dining club, recalls a parallel institution in another society. It may be defined with the help of the Lat. daps ‘banquet’. This word forms part of an etymological group which is well characterized in form, but has divergent meanings. Outside Latin, the root recurs in Greek dáptō (δάπτω) with the more general meaning “to devour,” but also in a nominal form which is closely connected with daps in spite of the apparent difference: dapánē (δαπάνη) ‘expenditure’. There are corresponding words in other languages: Old Icel. tafn ‘sacrificial animal, sacrificial meal’, Arm. tawn ‘festival’.

It will be noticed that all the forms have the same suffix –n. This formal feature brings in the Lat. damnum < *dap-nom, which is mentioned separately because at first sight it does not seem to be associated with the previous group.

Daps is a term of the religious vocabulary; the Scandinavian and Armenian expressions also belong to the same sphere. At an early date, within the historic period, daps had the sense of “banquet offered to the gods, festive meal.” The daps is described by Cato in De Agricultura with a characteristic expression of the old Latin religious vocabulary, dapem pollucere ‘to offer a sacred banquet’. This archaic expression pollucere is applied to the lavish feasts offered to the gods: polluctum.

Apart from this, there is evidence that daps is associated with the notions of abundance, lavish expenditure, generous offerings. Noteworthy are the adjective dapaticus and the adverb dapatice, obsolete forms collected and cited by Festus, who translates dapatice by magnifice ‘in a magnificent manner’. A verb dapino from daps or perhaps from the Greek dapanân, which had a closely related meaning, also existed. We have only a single example of dapino in Plautus (Capt., 897), but it is characteristic: aeternum tibi dapinabo victum ‘(if you tell the truth) I will offer you in perpetuity a sumptuous feast, I will entertain you royally for ever’.

A direct testimony defines the sense of daps, and dapatice as well as dapaticus confirm it: it is a “magnificent feast.” Ovid in the fifth book of the Fasti shows us a poor peasant to whom Jupiter appears in disguise. Suddenly, he reveals who he is: the peasant offers him as a daps his only possession, an ox, which he roasts whole. This is his most precious possession.

In Greek, dapanân means “to spend,” dapánē is “ostentatious expenditure.” In Herodotus the expression is applied to lavish expenditure. The adjectives Gr. dapsilḗs, Lat. dapsilis (coined on the Greek model) apply to what is abundant, ostentatious. Icel. tafn denotes the consumption of food; Arm. tawn, a solemn feast. From all this we may abstract a general notion, that of “expenditure on the occasion of a sacrifice which involves the consumption of large amounts of food,” expenditure required for a feast, for prestige, to display wealth.

We thus find in Indo-European a social phenomenon which in the language of the ethnologist is called potlatch; the display and consumption of wealth on the occasion of a feast. It is necessary to make a show of prodigality in order to demonstrate that one sets no store by it, to humiliate one’s rivals by the instantaneous squandering of accumulated wealth. A man conquers and maintains his position if he outdoes his rivals in this reckless expenditure. The potlatch is a challenge to others to do likewise in their turn. The competitors make a still more lavish outlay, and this results in circulation of wealth, which is accumulated and expended for the prestige of some and the enjoyment of others, as Mauss has shown so well.

In Indo-European there is no clear notion of rivalry; the agonistic character so prominent in archaic society has here a subordinate role. Nevertheless, emulation is not absent from this expenditure. In fact it is closely connected with hospitality (cf. daps and dapatice). We see the social roots of an institution which is a necessity in certain communities; its essence is the obligation to make a gift of food, which is understood to impose reciprocity. But these are ideas and terms of great antiquity, which are in the process of fading. In historic times there remains only damnum with the derived sense of “injury sustained, what is taken away by forcible seizure.” It is the expense to which one is condemned by circumstances or by certain legal stipulations. The peasant spirit and the legal exactitude of the Romans transformed the ancient conception: ostentatious expenditure became no more than an outright expenditure, what constitutes a loss. Damnare means to afflict a damnum on somebody, a curtailment of his resources; from this stems the legal notion of damnare ‘to condemn’.

Side by side with the words in which the ancient notion has survived, there are innovations which create a new concept, and this means that we have simultaneously two strongly contrasted aspects of an ancient idea.

The Hansa and Its Military Origins

Grain measurers depicted in their guild seat

Among the confraternities, where the participants in the communal banquet enjoy special privileges—those which characterize the guild in its medieval development—we encounter in the same economic and religious vocabulary of the Germanic world a close neighbor of ghilda, the hansa. This ancient term, which has survived down to modern times, designated in the countries around the North Sea an institution of great historic and economic importance. The Hansas are economic associations of groups of merchants; they constitute a society to which one belongs in virtue of a right which can be purchased, inherited or sold, and which forms part of commercial assets. The workings of this institution have been the object of numerous studies. The results of those who studied the origin of the word are negative: hansa has no certain etymology. Since no correspondence is found outside Germanic, it is the history in Germanic of the word which we must try to trace.

The story begins with the Gothic hansa, which gives a precise starting point to the analysis, although we have but few examples. In one passage, hansa translates, in apparently a vague way, the Greek plē̂thos ‘crowd’. But in three other examples, hansa corresponds to speîra (σπεῖρα) ‘cohort’. In Mark XV, 16: “the soldiers took Jesus inside the courtyard, that is to say into the pretorium, and they called together the whole cohort,” Got. alia hansa ‘totam cohortem’. It functions similarly in John XVIII, 3, 12. In the passage where plē̂thos is rendered by hansa (Luke VI, 17), if we read it in its entirety, we see that the translator had to translate successively ókhlos and plē̂thos. He chose hiuma ‘turba’ for ókhlos, and for plē̂thos ‘multitudo’ he used hansa ‘cohort’. This unit in fact comprised several hundred men, as many as a thousand, and could represent a “crowd” which in some way had been mobilized to welcome Jesus.

It is not by accident that hansa is found in Old High German in Tatian to translate cohors. In OE hōs is “the follower of a lord.” It is not until Middle High German that hans( e) assumes the sense of “commercial association” with the sense that it henceforward keeps. In Late Latin or in Latinized German hansa means a tax for a trade license as well as a commercial association.

The sense of “(military) cohort” indicates that one has to envisage the hansa as a company of warriors. Hansa would not have been employed in Gothic to translate speîra if it had, for instance, meant a religious group or a group with a common interest. In fact when Tacitus (Germania 13–14) describes the societies of young men (comitatus) which are attached to the chieftains, he gives us a picture of what the hansa must have been. These young men who attach themselves to a chieftain live from his bounty and receive abundant food which serves instead of pay (14, 4). They are always ready to follow him and defend him and to win renown under his orders.

It is probable that these companies of young warriors who vied with each other under their chieftain, while the chieftains competed among themselves to see who would attach to himself the keenest followers, formed the first model of the hansa. With the evolution of society, this company of warriors in which privileges and rites were shared was transformed into a society of companions of a different type, devoted to economic activities. The word remained but it was attached to a new reality.

Giving, Taking, and Receiving


1) Hittite, which attaches to the root *– the sense of “to take,” suggests that in Indo-European the notions “to give” and “to take” converged, as it were, in gesture (cf. English to take to).

2) Contrary to the traditional etymologies which find no difficulty in bringing together Lat. emo and Got. niman (Germ. nehmen), but firmly separate niman from Gr. némō, justifying both decisions by appeal to the sense, it can be shown that:

a) Got. niman and Gr. némō can be superimposed without difficulty on the basis of their original (technical) sense, which is preserved exactly in the Got. arbinumja and the Gr. klēronómos “heir.”

b) Lat. emo “take,” in its primitive gestural sense, has no etymological connection with Got. niman, which had originally a legal significance.


Hittite Empire Sphinx Gate, Alaca Höyük, Turkey / Photo by Bernard Gagnon, Wikimedia Commons

The expressions for “purchase” and “sale” are not separable from those for “give” and “take.” The root *– means “give” in all Indo-European languages. However, there is one language which fails to conform to this definition: in Hittite, – means “take” and pai– ‘give’. We cannot categorically affirm, given the inconsistent notation of Hittite consonants, that – really corresponds to the Indo-European *-; theoretically it could correspond to *dhē– ‘to place, put’, but this is not very likely. In general there is agreement in recognizing here— whatever the semantic evolution—the root *. The fact is that if we started from *dhē– to arrive at the sense “take” the semantic evolution would be even more obscure.

We simply have to take it as a fact that in Hittite – ‘take’ we have the contrary of the sense “give.” To explain this scholars have adduced as a parallel the form ā ‘take’ in Sanskrit. But here the preverb ā– is essential; it indicates movement towards the subject. With this preverb and the middle endings, the change to the sense “receive, take” is explicable within Sanskrit itself. Thus Sanskrit is of no direct help in explaining the sense of – in Hittite.

To explain it we may suppose that semantic shifts comparable to that undergone by the English word take in the expression to take to occurred within the ancient languages, but in different directions. This comparison may help us to discover the link between the two opposite meanings. Hittite and the other Indo-European languages have specialized in different ways the verb *-, which lent itself, according to the syntactical construction, to either sense. While Hittite – restricted the sense to “take,” the other languages constructed  with the idea of a destination, which results in the sense “to give.” [3]

This is not an artificial construction. Indo-European has several expressions for “take,” each of which specifies the notion in a different way. If one accepts that the original sense is that preserved in Hittite, the evolution leading to the meaning “give,” attested in the rest of the Indo-European domain, becomes intelligible.

Equally archaic is Hittite pai– “give.” It is explicable as a compound of the preverb pe– with *ai– ‘attribute, allocate’, a root attested in the Tokharian ai– ‘give’ and by several derived nouns, such as Av. aēta– ‘part’ and Osc. aeteis (gen. sing.), which translates Lat. partis.

The notion of “give” and “take” are thus linked in prehistoric Indo-European. It may be useful to consider in this connection an etymological problem relating to an already specialized word, Lat. emo, which, as will be shown below, once meant “take.” In another language a root is encountered with the same sense, which differs from the Latin form by the initial n-: this is Germanic *nem-, Got. niman, German nehmen ‘take’. Here we have two verbs of the same sense, Lat. em-, Germanic nem-; is there an etymological connection between them? This has often been accepted; but how can it be morphologically justified? Recourse is had to two devices: nem– may be composed of *(e)n + em or derived from a reduced form of ni + em. But in order to practice economy in reconstructions, we must first consider what matters most, although the least attention has been paid to it so far, i.e. the meaning.

The most ancient Germanic forms appear in Gothic. They are very frequent and instructive. The form niman presupposes *nem-, and we are acquainted with such a root. It appears in Gr. némō (νέμω), but the connection is ruled out because of the meaning of némō, which is not “take.” For the time being we do no more than point this out and turn our attention to niman. We have the simple verb as well as several compounds with numerous preverbs in various applications. The Greek verbs to which it corresponds are lambáneinaírein ‘take’, déksasthai ‘to receive’ (very frequent, especially in the expression “to receive grace”); the compounds with and– translate dékhesthai (apo-, para-); those with ga– (cf. German angenehm ‘pleasant’) “to receive, conceive, welcome” and also mente acciperematheîn ‘receive with the mind, learn’. There is a considerable preponderance of instances in which niman signifies not “take” but “receive.” In particular, a compound noun deserves attention because of its special technical meaning: arbi-numja ‘heir’. The first part, arbi, is an independent term which means “heritage,” Germ. Erbe, and which has considerable importance in the vocabulary of institutions. The form is clear: it is a neuter *orbhyom, which links up on the one hand with the Celtic terms of the same sense, e.g. Irl. orbe ‘heritage’, com-arbe ‘he who inherits’ (the connection is so close that here, as in many other cases, it is possible that this may be a borrowing by Germanic from Celtic). Another connection is with adjectival forms which may serve to throw light on the concept: Lat. orbus ‘bereft’, Arm. orb ‘orphan’, Gr. orpho-, orphanós. Outside Celtic the terms corresponding to arbi designate a person deprived of a parent, and also an orphan. The relationship between “heritage” and “orphan” may seem somewhat strange; but there is an exact parallel of meaning in another family of words. The Latin adjective hērēd– ‘heir’ has a certain correspondent in Greek in the agent noun khērōstḗs ‘collateral heir’ and also in the adjective khē̂ros ‘deprived of a parent’, fem. khḗra ‘widow’.

How can this etymological relationship be explained? In Homeric Greek, khērōstḗs is the member of the family who inherits in the absence of children, the relative who receives a property which has become “abandoned” (khē̂ros). Now in Gothic, arbi‘heritage’, derived from the neuter form *orbhyom, means literally ‘what devolves on the orbus’, that is to say, the property which is legally bestowed on a person who has suffered the loss of an immediate relative. It is the same idea as in hērēskhērōstḗs. According to Indo-European usage property is directly transmitted to the descendant, but he is not for this reason alone qualified as an “heir.” At that time, no need was felt for the legal precision which makes us qualify as “heir” the person who enters into possession of material wealth, whatever his degree of relationship with the deceased. In Indo-European, the son was not designated the “heir.” Heirs were only those who inherited in the absence of a son. This is the case with khērōstai, the collaterals who divided an inheritance where there was no direct heir.

Such is the relationship between the notion of “orphan, deprived of a relative” (son or father) and that of “inheritance.” It is illustrated by the definition given in a sentence from the Germania of Tacitus, Chapter 20: Heredesuccessoresque sui cuique liberi, et nullum testamentum, ‘everybody has as heirs and successors his own children, and there is no will and testament’; si liberi non sunt, proximus gradus in possessions fratres, patruiavunculi, ‘if there are no children it is the next of kin who enter into possession, the degrees of succession being brothers, paternal uncles and maternal uncles’.

Such are the arbi-numja. The literal sense of arbi-numja is “he who receives (numja) the heritage (arbi). We may now ask which Greek term arbi-numja translates? It is klēronómos (κληρονόμος). There is also an analytical expression arbi niman ‘to inherit’ which translates the Gr. klēronomeîn (κληρονομεῖν).

The formation of the Greek compound is instructive. The second term links up with némōnómosnomós, a very rich family of words which has been the subject of a study by E. Laroche (Histoire de la racine nem- en grec ancien, 1949), in which its uses are examined in detail. This extremely important root has a rich variety of derivatives. The notion which is elicited is that of a legal division or sharing out, exclusively enjoined by law, custom, or by agreement, but not by arbitrary decision. Other verbs in Greek mean “divide”; an example is datéomai, but here the difference is this: némō is “to divide according to agreement or the law.” It is for this reason that pastureland which has been shared out according to customary law is called nomós. The meaning of nómos ‘the law’ goes back to “legal apportionment.” Thus némō is defined in Greek as “to divide legally” and also “to acquire legally by way of apportionment” (this being the sense of the active).

Is it an accident that the Gothic (arbi-)numja has the same formation as (klēro-)nómos, seeing that there would be no occasion to use the verb niman to translate kléronomeîn if it meant “to take” ?

We can now see how the correspondence in a technical sense is arrived at between némō and niman: it is because Gothic niman means “take” in the sense of “receive legally” (cf. the use in which it corresponds to the Greek dékhesthai); hence comes the sense “receive, receive one’s share, take.” We may consider this expression arbi niman and the compound arbi-numja ‘heir’ as one of those where the ancient meaning of niman survives, the same meaning which némō had in Greek and which led to the formation of the term klēronómos ‘heir’. The other usages are easily explicable. [4]

Thus the Germanic niman has nothing to do with emo. We must postulate a Germanic root nem– which, in the light of this interpretation of its sense, links up with the group of Indo-European forms from the root *nem-, which are also abundantly represented in Greek.

To what result do we come if we subject emo to like scrutiny? Correspondences with initial e– are found in Old Slavonic imǫ, and in Baltic in the Lith. imùim̃ti “take.” Latin helps to delimit the meaning of emo, which is “to draw back, to take away.” Eximois to “take out of,” while the meaning of eximius corresponds in sense to Gr. éxokhos ‘outstanding, preeminent’. Further, we have exemplum which, by a curious development, means “an object set apart, separated by its very marked characteristics,” hence “model, example”; prōmo means “draw from (a store)” and its verbal adjective promptus ‘taken out, drawn, ready to hand’. Per-imo (with the meaning of the preverb which we find in per-do) means “make disappear, annihilate”; sūmo (from *subs-emo) ‘take by lifting’.

All this shows that the Latin sense “take < draw, remove, seize” has no connection with “take < receive, welcome” of Germanic. These are quite different notions in origin, and they reveal their peculiarity if we succeed in grasping their first sense. Each of them has its own domain and history. It is only at the end point of their evolution and in the most watered-down sense that Germanic niman and Latin emo resemble each other.

We return to emo ‘buy’. The manner in which emo develops a restricted sense in Latin suggests that the meaning “buy” implies a quite different conception from that inherent in the terms belonging to the Greek family of pérnēmi, etc. It is clear that emo at first meant “take to oneself, draw to oneself.” The possession which it affirms is expressed by the gesture of the man who takes the object and draws it to himself. The sense of “buy” must first have evolved with reference to human beings whom one “takes” after having fixed a price. The notion of “purchase” had its origin in the gesture which concluded the purchase (emo) and not in the fact of paying a price, handing over the value of the object. [5]



In Latin “guest” is called hostis and hospes < *hosti-pet-. What is the meaning of these elements? What is the meaning of the compound?

1) –pet-, which also appears in the forms pot-, Lat. potis (Gr. pótisdespótēs, Skr. patiḥ), and –pt– (Lat. –pteipse?) originally meant personal identity. In the family group (dem-) it is the master who is eminently “himself” (ipsissimus, in Plautus, means the master); likewise, despite the morphological difference, Gr. despótēs, like dominus, designated the person who personified the family group par excellence.

2) The primitive notion conveyed by hostis is that of equality by compensation: a hostis is one who repays my gift with a counter-gift. Thus, like its Gothic counterpart, gasts, Latin hostis at one period denoted the guest. The classical meaning “enemy” must have developed when reciprocal relations between clans were succeeded by the exclusive relations of civitas to civitas (cf. Gr. xénos ‘guest’ > ‘stranger’).

3) Because of this Latin coined a new name for “guest”: *hostipet-, which may perhaps be interpreted as arising from an abstract noun hosti “hospitality” and consequently meaning “he who predominantly personifies hospitality, the one who is hospitality itself.”

The study of a certain number of expressions relating to exchange, especially those based on the root *mei-, like the Latin mūnus ‘an honorific post implying an obligation to reciprocate’, I.-Ir. Mitra, the personification of a reciprocal contract (as illustrated in Iliad VI, 120–246), *meit– in the Latin mūtuus, Skt. mithu– ‘changed (falsely)’ > ‘lie’, Av. miθwara ‘pair’, also leads us to a word for “guest”: mēhmān in middle and modern Iranian. Another word for “guest” in modern Iranian, ērmān < aryaman, links up with a very special kind of “hospitality” within a group of the Arya, one of the forms of which is reception by marriage.


From the earliest times to the present, slavery has existed in various forms.

The vocabulary of Indo-European institutions throws up some important problems, the terms of which have, in some cases, not yet been posed. We become aware of their existence and even partly create the object of our study by examining words which reveal the existence of an institution, the traces of which we can barely glimpse in the vocabulary of this or that language.

One group of words refers to a well-established social phenomenon, hospitality, the concept of the “guest.” The basic term, the Latin hospes, is an ancient compound. An analysis of its component elements illuminates two distinct notions which finally link up: hospes goes back to *hosti-pet-s. The second component alternates with pot-, which signifies “master,” so that the literal sense of hospes is “the guest-master.” This is a rather peculiar designation. In order to understand it better we must analyze the two elements *potis and hostis separately and study their etymological connections.

The term *potis first merits a brief explanation in its own right. It presents itself in its simple aspect in Sanskrit pátiḥ ‘master’ and ‘husband’ and in Greek posís ‘husband’, or in composition as in despótēs.

In Sanskrit the distinct senses “master” and “husband” correspond to different declensions of one and the same stem; but this is a development peculiar to Sanskrit. As for Gr. posís, a poetical word for “husband,” it is distinct from despótēs, where the sense “master of the house” is no longer felt; despótēs is solely an expression of power, whereas the feminine déspoina conveys the idea of “mistress,” a title of majesty.

The Greek term despótēs, like the Sanskrit correspondent dám pátiḥ, belongs with a group of ancient compound words, each of which had as its first element the name of a social unit of variable extension:

dám pátiḥ (master of the house)
viś „ (master of the clan)
jās „ (master of the “lineage”)

Apart from despótēs and dám pátiḥ, the only one attested in a number of languages is the compound which is in Sanskrit viś-pátiḥ and in Lithuanian vë̃š-pats ‘clan chief’.

In Latin an extensive word family is organized around the word *potis either as a free form or in composition. Apart from hospes it forms the adjectives imposcompos ‘who is not…’ or ‘who is master of himself, of his senses’ and the verb *potire, the perfect of which, potui, survives incorporated into the conjugation of the verb meaning “be able,” possum, which itself is formed from the adjective potis in a predicative use: potis sumpote est, an expression which is simplified to possum, potest.

All this is clear and there would be no problem, the sense being constant and the forms superimposable, had not *potis at two points of the Indo-European world developed a very different sense. In Lithuanian it provides the adjective pats ‘himself’ and also the substantive pats ‘master’ (in composition vë̃š-pats). Parallel to this, we find in Iranian the compound adjective x u aē-paiθya ‘one’s own’, ‘of oneself’, which is used without distinction of person: “mine, yours, his”; “one’s own.” xu is an Iranian form of the ancient reflexive pronoun *swe, *se, literally “of oneself,” and –paiθya derived from the ancient *poti-. These facts are well known, but they deserve careful scrutiny because of the singularity of the problem which they pose. Under what conditions can a word denoting “master” end up signifying identity? The primary sense of *potis is well defined, and it had a strong force: “master,” whence in marriage “husband,” or in social terminology the “chief of some unit, whether house, clan, or tribe.” But the sense “oneself” is also well attested. Here Hittite makes an important new contribution. It offers no form corresponding to *potis, whether as adjective or substantive. Despite the early date at which it appears, Hittite has a vocabulary which has already been transformed to a considerable extent. Many notions now are conveyed by new terms. The interesting point in the present connection is that Hittite presents an enclitic particle, –pet (-pit), the sense of which is “precisely (him)self,” a particle of identity referring to the object under discussion. An example is the following:

takku IR-iš huwāi

naš kururi KURe paizzi

kuišan EGIR-pa uwatezzi

nanzan apāšpit dai.

If a slave flees,

and if he goes to an enemy country,

the one who brings him back,

that very one takes him.

In this demonstrative apāš-pit ‘that one precisely, that very one’, the particle –pit establishes a relation of identity. It has, incidentally, the same function whether attached to a demonstrative, a noun, or even a verb. It is evident that the use of this particle corresponds to the sense of identity of *potis found in Lithuanian and in Iranian.

Once the sense, the form and the use is established in these languages, we discover elsewhere other forms which can be linked with them in all probability. The Lithuanian particle pat signifies “exactly, precisely,” like the Hittite –pet. With this may be compared Lat. utpote, the analysis of which must be rectified. It does not mean etymologically “as is possible” (with the pote of pote est) but “precisely inasmuch,” with pote marking the identity. Utpote emphatically identifies the action with its agent, the predicate with the person who assumes it. We may also add the Latin postposition –pte in suopte (Festus: suopte pro suo ipsius‘his very own, what belongs to that very person’). A further example, but this is less certain, is the mysterious –pse of ipse. In any case, if we confine ourselves to the two Latin facts and to the Lithuanian pat, we can establish the survival of a use of *pot– to designate the person himself, and to assign to him the possession of a predicate affirmed in the sentence. Accordingly, what was considered as an isolated use becomes an important indication and reveals to us the proper signification of potis. While it is difficult to see how a word meaning “the master” could become so weakened in force as to signify “himself,” it is easy to understand how an adjective denoting the identity of a person, signifying “himself,” could acquire the sense of master. This process, which illustrates the formation of an institutional concept, can be corroborated elsewhere: several languages have come to designate “the master” by a term meaning “himself.” In spoken Latin, in Plautus, ipsissimus indicates the “master (mistress), the patron,” the (personage) himself, the only one who is important. In Russian, in peasant speech, sam ‘himself’ refers to the “lord.” Among a restricted but important community, the Pythagoreans, the formula autòs éphā (ἀυτὸς ἔφα) ‘he himself has said it’, with autós referred to the “master” par excellence, Pythagoras, was used to specify a dictum as authentic. In Danish, han sjølv ‘er selbst’ has the same meaning.

For an adjective meaning “himself” to develop into the meaning “master” there is one necessary condition: there must be a circle of persons subordinated to a central personage who assumes the personality and complete identity of the group to such an extent that he is its summation: in his own person he is its incarnation.

This is exactly the development we find in the compound *dem-pot(i)- ‘master of the house’. The role of the person so named is not to give orders but to assume a representation which gives him authority over the family as a whole with which he is identified.

A verb derived from *poti-, like Skt. pátyate, Lat. potior ‘to have power over something, have something at one’s disposal’, already marks the appearance of a sense of “to be able to.” With this may be compared the Latin verb possidēre ‘possess’, stemming from *pot-sedēre, which describes the “possessor” as somebody who is established on something. The same figurative expression has passed into the German word besitzen. Again, in Latin we have the adjective compos ‘he who is master, who has command of himself’. The notion of “power” (theoretical) is thus constituted and it receives its verbal form from the predicative expression pote est, contracted to potest, which gives rise to the conjugation possumpotest ‘I am capable, I can’. [6]

It is worthwhile pausing for a moment to consider a peculiar fact: as against Skt. dam pati and Gr. despótēs, Latin has formed from the same root an equivalent expression, but by a different procedure: this is dominus, a secondary derivative which belongs to a series of expressions for “chief.” Thus tribunus ‘chief of the tribe’, in Gothic kindins (< *genti-nos) ‘chief of the gens’; *druhtins (OHG truhtin) ‘chief of the body’; þiudans < *teuta-nos ‘king, chief of the people’. This morphological process, whereby *-nos is suffixed to the name of a social unit, has furnished in Latin and Germanic expressions for chiefs of political and military groups. Thus, by independent paths, the two series link up: on the one hand by means of a suffix, on the other by a compound word, the term for the master has been coined from the social unit which he represents.

We must return now to the compound which provoked this analysis, hospes, this time in order to study the initial term, hostis. Among the expressions common to the prehistoric vocabulary of the European languages it is of special interest: hostis in Latin corresponds to gasts of Gothic and to gostĭ of Old Slavonic, which also presents gos-podi ‘master’, formed like hospes.

But the meaning of Gothic gasts and OSl. gostĭ is “guest,” whereas that of Latin hostis is “enemy.” To explain the connection between “guest” and “enemy” it is usually supposed that both derived their meaning from “stranger,” a sense which is still attested in Latin. The notion “favorable stranger” developed to “guest”; that of “hostile stranger” to “enemy.”

In fact, “stranger, enemy, guest” are global notions of a somewhat vague character, and they demand precision by interpretation in their historical and social contexts. In the first place, the signification of hostis must be narrowed down. Here we are helped by the Latin authors themselves who furnish a series of words of the same family and also some instructive examples of the use of the term hostis. It preserved its ancient value of “stranger” in the law of the Twelve Tables, e.g.: adversus hostem aeterna auctoritas est(o), no word of which, with the exception of the verb “to be,” is employed in the same sense as in classical Latin. It must be understood as “vis-à-vis a stranger, a claim for property persists forever,” that is, it never lapses when it is against a foreigner that the claim is introduced. Of the word hostis itself, Festus says: eius enim generis ab antiquis hostes appellabantur quod erant pari iure cum populo Romanoatque hostire ponebatur pro aequare ‘in ancient times they were called hostesbecause they had the same rights as the Roman people, and one said hostire for aequare’. It follows from this note that hostisis neither the stranger nor the enemy. We have to proceed from the equivalence of hostire = aequare, while the derivative redhostire is glossed as referre gratiam ‘repay a kindness’ in Festus. This sense of hostire is still attested in Plautus: Promitto hostire contra ut merueris ‘I promise you a reciprocal service, as you deserve’ (Asin. 377). It recurs in the noun hostimentum, explained as beneficii pensatio ‘compensation of a benefit’, and also aequamentum ‘equalization’. To a more specialized technique belongs hostus, an archaic term of the language of agriculture, cited and explained by Varro, R.R. 1, 24, 3: hostumvocant quod ex uno facto olei reficitur ‘one calls hostus the amount of oil obtained in a single operation of the press’. In some way the product is considered as a counterpart. Another technical term is hostorium, a stick for use with a bushel measure so as to keep a constant level. The old Roman pantheon, according to S. Augustine, knew a Dea Hostilina, who had as her task to equalize the ears of corn or to ensure that the work accomplished was exactly compensated by the harvest. Finally, a very well-known word, hostia, is connected with the same family: its real sense is “the victim which serves to appease the anger of the gods,” hence it denotes a compensatory offering, and herein lies the distinction which distinguishes hostia from victima in Roman ritual.

It is a striking fact that in none of these words, apart from hostis, does the notion of hostility appear. Primary or derived nouns, verbs or adjectives, ancient expressions of the religious language or of rural vocabulary, all attest or confirm that the first sense is aequare ‘compensate, equalize’.

How does hostis itself fit in with this? This emerges from the definition of Festus already cited: “quod erant pari iure cum populo Romano.” This defines the relation of hostis and hostire: “the hostes had the same rights as the Romans.” A hostis is not a stranger in general. In contrast to the peregrinus, who lived outside the boundaries of the territory, hostis is “the stranger insofar as he is recognized as enjoying equal rights to those of the Roman citizens.” This recognition of rights implies a certain relation of reciprocity and supposes an agreement or compact. Not all non-Romans are called hostis. A bond of equality and reciprocity is established between this particular stranger and the citizens of Rome, a fact which may lead to a precise notion of hospitality. From this point of view hostis will signify “he who stands in a compensatory relationship” and this is precisely the foundation of the institution of hospitality. This type of relationship between individuals or groups cannot fail to invoke the notion of potlach, so well described and interpreted by Marcel Mauss in his monograph on “le Don, forme primitive de ľéchange,” Année sociologique, 1924. This system which is known from the Indians of Northwest America consists of a series of gifts and counter-gifts, each gift always creating an obligation of a superior gift from the partner, in virtue of a sort of compelling force. It is at the same time a feast connected with certain dates and cults. It is also an economic phenomenon, insofar as it secures circulation of wealth; and it is also a bond between families, tribes and even their descendants.

The notion of “hospitality” is illuminated by reference to potlach, of which it is a weakened form. It is founded on the idea that a man is bound to another (hostis always involves the notion of reciprocity) by the obligation to compensate a gift or service from which he has benefited.

The same institution exists in the Greek world under a different name: xénos (ξένος) indicates relations of the same type between men bound by a pact which implies precise obligations that also devolve on their descendants. The xenía (ξενία), placed under the protection of Zeus Xenios, consists of the exchange of gifts between the contracting parties, who declare their intention of binding their descendants by this pact. Kings as well as private people act in this way: “(Polycrates) had concluded a xenía (with Amasis) and they sent each other presents” ξενίην συνεθήκατο (verb of making a compact) πέμπων δῶρα καί δεκόμενος ἄλλα παρ’ ἐκείνου (Herodotus III, 39). Mauss (Revue des Etudes grecques, 1921) finds an example of the same custom among the Thracians. Xenophon wanted to conclude arrangements for the food supplies of his army. A royal councilor tells him that if he wants to remain in Thrace and enjoy great wealth, he has only to give presents to King Seuthes and he would give him more in return (Anabasis VII, 3; X, 10). Thucydides (II, 97) gives much the same testimony apropos of another Thracian king, Sitalkes: for him it is more shameful not to give when one is asked to do so than not to receive when one has asked. In the civilization of Thrace, which seems to have been rather archaic, this system of obligation was still preserved in its full force.

One of the Indo-European expressions of this institution is precisely the Latin term hostis, with its Gothic correspondent gastsand Slavic gospodĭ. In historical times the custom had lost its force in the Roman world: it presupposes a type of relationship which was no longer compatible with the established regime. When an ancient society becomes a nation, the relations between man and man, clan and clan, are abolished. All that persists is the distinction between what is inside and outside the civitas. By a development of which we do not know the exact conditions, the word hostis assumed a “hostile” flavor and henceforward it is only applied to the “enemy.”

As a consequence, the notion of hospitality was expressed by a different term in which the ancient hostis nevertheless persists, but in a composition with *pot(i)s: this is hospes < *hostipe/ot-s. In Greek, the guest (the one received) is the xénos and he who receives is the xenodókhos (ξενοδόχος). In Sanskrit, atithi ‘guest’ has as its correlate atithi-pati ‘he who receives’. The formation is parallel to that of Latin hospes. The one who receives is not the “master” of his guest. As we have seen, –pot– did not have originally the meaning of “master.” Another proof of this is the Gothic brūþ-faþs ‘newly married man, νύμφιος’, the German equivalent of which is Bräutigam ‘bridegroom’. From bruþ ‘newly married woman’ was created the corresponding designation for the “newly married man,” either with *potis as in Gothic brūþ-faþs, or with guma ‘man’, like in the German Bräutigam.

The formation of *ghosti– (hostis) deserves attention. It looks like an abstract word in –ti which has become a personal qualification. All the ancient compounds in –poti– have in effect as their first element a general word designating a group: thus *dems-potijās-pati. We thus understand better the literal sense of *ghosti-petshospes as the incarnation of hospitality. In this way we link up with the above definition of potis.

Thus the history of hostis recapitulates the change brought about in Roman institutions. In the same way xénos, so well characterized as “guest” in Homer, later became simply the “stranger,” the non-national. In Attic law there is a graphḕ xenías, a lawsuit against a “stranger” who tries to pass for a citizen. But xénos did not evolve the sense of “enemy” as did hostis in Latin.

The semantic mechanism described for hostis has a parallel in another order of ideas and another series of words. It concerns those which come from the root *mei– ‘exchange’, Skt. ni-mayate ‘he exchanges’ and especially the Latin term mūnus (< *moi-nos, cf. the archaic form moenus). This word is characterized by the suffix –nes, the value of which was determined by Meillet (Mem. Soc. Ling., vol. XVII) in pignusfacinusfūnusfēnus, all words which, like mūnus, refer to notions of a social character; cf also Skt. rek-naḥ ‘heritage’, etc. In fact mūnus has the sense of “duty, a public office.” From it are derived several adjectives: mūnisimmūniscommunis. The last has a parallel in Gothic: ga-mains, German gemein ‘common’.

But how can the notion of “charge, responsibility, public office” expressed by mūnus be associated with that of “exchange” indicated by the root? Festus shows us the way by defining mūnus as “donum quod officii causa datur” (a gift made for the sake of an officium). In fact, among the duties of a magistrate mūnus denotes spectacles and games. The notion of “exchange” is implied by this. In nominating somebody as a magistrate one confers on him honor and certain advantages. This obliges him to render counter-service in return, in the form of expenditure, especially for games and spectacles. In this way we can better understand the affinity between gratus and mūnis (Plautus, Merc. 105), and the archaic sense of immūnis as “ingratus” (that is to say, one who fails to make due return for a received benefit). If mūnus is a gift carrying the obligation of an exchange, immūnis is he who does not fulfill his obligation to make due return. This is confirmed in Celtic by Irl. moin (main) ‘precious objects’, dag-moini ‘presents, benefits’. Consequently communis does not mean “he who shares the duties” but really “he who has munia in common.” Now if the system of compensation is active within one and the same circle, this determines a “community,” a group of persons united by this bond of reciprocity.

Thus the complex mechanism of gifts which provoke counter-gifts by a kind of compelling force finds one more expression among the terms derived from the root *mei-, like mūnus. If we did not have the model of this institution, it would be difficult to grasp the meaning of the terms which refer to it, for it is within this precise and technical framework that these terms find their unity and proper relations.

A further question now arises: is there no simple expression for “gift” which does not call for a return? The answer is already given. It emerges from a previous study: there exists an Indo-European root, that of Latin dodōnum, Greek dō̂ron. It is true, as we have seen above (Book One, Chapter Five), that the etymological prehistory of *– is by no means straightforward but is a criss-cross of apparently contradictory facts.

Nevertheless, in historical times the notion of “give” is everywhere attached precisely to the form of *-, and in each of the languages (except Hittite) it gives rise to parallel formations. If in Greek the term down does not indicate in itself and unequivocally “gift” without reciprocity, the meaning of the adverb doreán ‘gratuitously, for nothing’ is sufficient guarantee that the “gift” is really a disinterested one. We must further mention forms stemming from another root which is little known and represented but which must be re-established in its importance and antiquity: this is the root *ai-. From it is derived the verb ai-tsi ‘give’ in Tokharian, as well as the Hittite pai– (formed by attachment of the preverb pe– to ai-) ‘give’. Greek has preserved a nominal form aîsa (αἶσα) ‘lot, share’. In Oscan an abstract *ai-ti– ‘part’ is attested by the genitive singular aeteis, which corresponds in meaning to the Latin genitive partis. Finally, Illyrian onomastics presents us with the proper name Aetor, which is the agent noun from this same root ai-. Here we have evidence for a new expression for “give” conceived as “assigning a portion.”

Returning now to the words belonging to the etymological family represented in Latin by mūnusimmūniscommunis, we can pick out in Indo-Iranian a derivative of considerable importance and peculiar formation. This is a divine personification, the Indo-Iranian god Mitra, formed from *mei-, in a reduced form, with the suffix –tra-, which generally serves to form the neuter nouns for instruments. In Vedic, mitra– has two genders, masculine as the name of the god and neuter in the sense of “friendship, contract.” Meillet, in a famous article (Journal Asiatique, 1907) defined Mitra as a divinized social force, as the personified contract. But both “friendship” and “contract” may be given further precision by siting them in their context: what is concerned is not sentimental friendship but a contract in so far as it rests on an exchange. To make clear these notions as they were practiced and lived in ancient society, we may recall a Homeric scene which gives what might be called a “sociological” illustration. It is the celebrated episode of the sixth book of the Iliad, lines 120–236.

Glaucos and Diomedes Exchange Armour pélikè attique du Peintre de Hasselmann, v.  420 av. J.-C / Wikimedia Commons

Glaucus and Diomedes, face to face, are trying to identify each other and discover that their fathers are bound by the bonds of hospitality (174). Diomedes defines his own position vis-à-vis Glaucus:

Yes, you are for me an hereditary guest (xeînos) and that for a long time (215)…thus I am your host in the heart of the Argolid and you are mine in Lycia, the day when I shall go to that country. From now on we shall both avoid each other’s javelin (224–226)…Let us rather exchange our weapons so that everyone may know here that we declare ourselves to be hereditary guests (230–231).

This situation gives each of the contracting parties rights of greater force than the common national interest. These rights are in principle hereditary, but should be periodically renewed by means of gifts and exchanges so that they remain personal: it is for this reason that the participants propose to exchange arms. “Having thus spoken, they leap from their chariots, take each other by the hand and pledge their faith. But at that moment Zeus…stole away Glaucos’ reason because in exchanging arms with Diomedes…he gives him gold in exchange for bronze, the value of one hundred oxen in exchange for nine” (232–236).

Thus the bard sees here a fool’s deal. In reality the inequality of value between the gifts is intentional: one offers bronze arms, the other gives back arms of gold; one offers the value of nine oxen, the other feels himself bound to render the value of one hundred head of cattle.

This episode serves to throw light on the manifestations which in this society accompany the type of engagement which we call a “contract,” and to restore its proper value to a term like Skt. mitra-. Such is the mitra– between Diomedes and Glaucus, an exchange which is binding and contractual. It also makes clear the formal analysis of the term. This suffix –tra– may form an agent noun as well as an instrumental one, the grammatical gender varying according to whether the action is the work of an instrument or a man: hence we have along with the neuter mitram, the masculine mitras. We might examine mythology and try to discover in the role of Mitra the survivals of its etymological origin. But first we must extend the inventory of notions which were formed from the same root and which are related to those which we have been studying. Closely related to *mei– is a form *mei-t– with the suffix –t-, which appears in the Latin verb mūtō ‘change’, ‘exchange’. The signification may be more precisely delimited if it is compared with the adjective mūtuus ‘reciprocal, mutual’. We must also consider a particular use of the adjective: mūtua pecūnia ‘money lent or borrowed’, as well as the verb derived from the adjective as thus used, mūtuāre ‘borrow’, i.e. to take money with the obligation to repay it. Thus “loan” and “borrowing” enter in their turn into the cycle of exchange. This is not the end of the matter. “Exchange” here has a close affinity with the “gift.” The Gothic correspondent of the Latin from mūtōmūtuus is maidjan ‘exchange’. Now the derived noun maiþms (from *mait-mo-) translates the Greek dō̂ron ‘gift’, but in a passage where it implies “recovery” and to a certain extent “exchange.”

The other derivatives are divided into:

1) one group with a specialized sense, e.g. Skt. mithu– ‘false, lie’; as with Latin mūtō, the idea of “changing” leads to that of “altering.” When we say of somebody that he has altered, this is rarely to his advantage.

2) A series of other derivatives, however, preserve the proper sense. This is particularly so in Iranian: e.g. Avestan miθwara– ‘paired’; maēθman– < *mei-t-men ‘pairing’. A development of a social character gives to maēθman the sense of “mutuality,” and this leads to the designation of the “guest” in Middle and Modern Iranian by mēhmān < *maēθmānam (accusative), which by a long detour brings us back to our starting point. Once again we end up by defining the “guest” by the notion of mutuality and the bonds of reciprocity. [7]

There is another term for the “guest” in modern Iranian: ērmān, the ancient form of which is attested as aryaman ‘intimate friend’, a term well known in Indo-Iranian. This is also the name of a mythological figure, the name of a god. Aryaman is the god of hospitality. In the Rig Veda, as in the Atharva, he is especially associated with marriage.

In whatever way we interpret the formative –man (this must be a nominal form), the name of the god Aryaman is connected with the term arya. We shall see later in this work that arya is the common and reciprocal term used by members of a community to designate themselves. It is the name for a man of the same language and the same race. This explains why one of Aryaman’s functions was to admit individuals into an exogamic community, called “Aryan,” through a marriage ceremony: it is a kind of internal hospitality, a tribal alliance. Aryaman intervenes when a woman taken from outside the clan is introduced for the first time as a wife into her new family.

Aryaman later came to be used in a number of different senses. The Persian ērmān ‘guest’ has been quoted above. In the language of the Ossetes, an Iranian people occupying an enclave in the Caucasus with institutions and vocabulary of great antiquity, the word limän means “friend,” and this is the regular phonetic development of aryaman. The bonds of relationship, of family and tribal friendship, are redefined in each language accordingly as the terminology remains fixed or evolves. These terms, far removed from one another, came back to the same problem, that of institutions of welcoming and reciprocity, thanks to which the men of a given people find hospitality in another, and whereby societies enter into alliances and exchanges. We have found a profound relationship between these institutional forms as well as a recurrence of the same notions behind a terminology which is sometimes refashioned.

Personal Loyalty


For Osthoff, Eiche und Treue (1901), the group of Germ. treu is related to the Indo-European name for “oak,” Gr. drûs: to be loyal means to stand as firm as an oak. It will be shown that if the relationship really exists, the affiliation is the reverse: the common root signifies “to be firm” and the adjective designates “tree,” literally “what is resistant, the solid one” (the meaning of “oak” is limited to a period of Greek and should not be attributed to the time of Indo-European unity).

Between Germanic *drauhti– (Got. gadrauhts ‘soldier’) and *drauhtino– (old Icelandic drottin– ‘chief lord’), the affiliated words in Slavic and Baltic meaning “friend, companion” allow us to establish the link known elsewhere (in dominustribūnus, etc.) between the nominal expression and its derivative in –no-. *drauhti is a collective designating “company” (in the military sense, as described for us by Tacitus, Germ. 13) and drauhtino-, the princeps who impersonates authority.

In the light thrown both by the Germanic legends concerning Odin Herjan and by Tacitus Germania 43, Gothic harjis (Germ. Herr) is revealed as the name of a group of masqueraders who on occasion assembled for plundering expeditions. (Although Gr. koíranos may formally correspond to herjan, the meaning which emerges from Homeric usage prompts the rejection of this purely formal equation.)

Lat. fidēs preserves a very ancient meaning, blurred and simplified in other languages where the root *bheidh is represented, and altered even in Latin itself after a certain period; its meaning was not “trust” but “the inherent quality of a person which inspired confidence in him and is exercised in the form of a protective authority over those who entrust themselves to him.” This notion is very close to that of *kred– (studied below in Chapter 15). So we can understand why Lat. fidēs was at all periods the noun corresponding to credo.


The terms which we have studied up to now have all been concerned with the relationships of man to man, in particular the notion of “hospitality.” From this point of view, which is both personal and institutional, we shall now consider the notion of personal loyalty within a particular group of languages, but with reference to the common Indo-European vocabulary: that is to say, the bond established between a man who possesses authority and the man who is subjected to him by a personal pledge. This “loyalty” gives rise to an institution which is very ancient in the western Indo-European world and which is most clearly apparent in the Germanic world.

Part I

The designation of this concept appears in an expression represented today by the German Treue and which is well attested in all Germanic dialects: in Gothic by the verb (ga)trauan, which translates πεποιθέναι ‘to have faith’, the noun trauains, πεποίθησις, ‘trust’, trūa in Icelandic, truōn in Old English (German trauen), all derived from a nominal stem *truwō; Icelandic trū‘respect, trust bestowed’, from which is derived Icelandic trur ‘loyal, faithful’. The action noun derived from this root has undergone a considerable development and has persisted for a long time in Germanic vocabulary: Gothic trausti ‘pact, alliance’, which translates διαθήκη, Icelandic traustr ‘reliable, sure, loyal’.

This is the source of the modern derivatives some of which designate a pact of alliance, an agreement, the pledged word, while others, verbs and nouns, have the meaning to “inspire confidence,” to “reassure,” to “console”; on the one hand we have the group represented by the English “trust” and on the other the group represented by the German trösten ‘console’. These moral notions are clearly bound up with an institution. In Germanic feudal vocabulary the Latinized form trustis designates the bond of fealty and also those who have thus bound themselves and who form the followers of a personage. The Old High German noun Traue is the source of the French trève ‘truce’.

The diversity of the Germanic forms shows the complexity of this idea, which results in terms as different as Germ. Treuetrauen ‘to have trust’, Trost ‘consolation’, Engl. trusttrue and truce. They all have one and the same origin in a Germanic root *dreu-, from which stems a Germanic abstract *drousto– (Old Icel. traust ‘faith, trust’, Germ. Trost ‘consolation’), a derivative *draustyo– (Gothic trausti ‘pact’) and an adjective *dreuwo– (Gothic triggws ‘faithful’, German treu).

This group of words was studied by the etymologist H. Osthoff, in his Etymologica Parerga (1901), a collection of different etymological studies, one chapter of which is entitled “Eiche und Treue” (‘Oak and Loyalty’). This strange title summarizes the substance of a lengthy study (about a hundred pages) which starts with this word family and connects it up with an Indo-European prototype, which he thought was the name of the “oak.” The formal basis of the deduction is a connection of the Indo-European *dreuwo with Greek drū̂s (δρῦς) ‘oak’. Osthoff considers that the “oak,” the hardiest and strongest of the trees, was the symbol of qualities the most abstract expression of which is found in this group of words with reference to the notion of “loyalty.” Thus the “oak” on this showing stood as a symbol of institutional “loyalty.” This demonstration has found a place in our etymological dictionaries, so it is important to check its foundations. Every etymological reconstruction must give the greatest weight to the dialect distribution of the forms and to the relationships which emerge from them in the classification of the different senses. Now it can be shown that Osthoff’s study completely falsifies the whole history of these terms; the true relations of the facts have been reversed.

In effect, if Osthoff is right, the name of the oak should be a common Indo-European one: it must have existed in all languages and in the given sense. We should thus expect to find a primary term in Indo-European, of constant form and sense, designating the “oak.” This is far from being the case. This word for “oak” appears only in one language and only at a certain period of that language. Before we begin to discuss it at all, one point of fact must be made. The oak is a tree limited to a specific area. The Indo-Europeans could not have known and designated it with a common name because it does not exist everywhere: there is no word for oak in Indo-Iranian for a very good reason. It is a tree of Central Europe and only the languages of Central and Eastern Europe have a word to denote it.

It would appear that this lexical distribution corresponds to the movement of the Indo-European peoples towards their historical sites. Everything—the historical, linguistic and archaeological facts—indicates that migration took place from east to west and that the Germanic peoples were among the last to be installed in the regions which they now occupy. This migration took place in several stages along a route which we can work out, and it ended in the region where the oak is found. It certainly did not start from that region.

This is confirmed by an examination of the names for the oak. The Indo-European form appears in two guises *de/orw– and *drew– with a full and a reduced degree respectively of the root and of the suffixal element, conforming to the well-established pattern of the Indo-European root. From these two forms came respectively the Gr. dóru (δόρυ) and drûs. In studying the senses, we shall take together the forms which derive from one or the other form of the root. Now it can be seen that the radical *dreu– with its alternative forms *drū-, *doru– exclusively designates “tree.” Thus Gothic triu translates Gr. xúlon ‘tree, wood’, and this is the sense in most languages. It is easy to establish that the old Slavic druva signifies “wood,” that the Indo-Iranian forms drūdāru denote exclusively “tree,” “wood” and “plant.” In the Avestan material the adjective drvaēna, like the Gothic triweinswhich corresponds to it, is applied to a “wooden” object. In certain languages a secondary differentiation between the derivates took place, such as in Old Slavic between drevo ‘tree’ (from *derwo-) and druva ‘wood’ (from druwo).

The Greek forms are of particular interest in this connection. From the same root Greek has derived two historically distinct, but evidently related, terms: dóru ‘(wood of) the spear’ and drū̂s ‘oak’, which we must consider in greater detail. The first sense of dóru is “tree, sapling”; thus in Od. 6, 167 Odysseus says to Nausicaa: “I have never seen grow from the earth such a tree (dóru).”

It is also the wood used in the construction of ships: δόρυ νηΐον, the keel of a ship; further, it is the “wood” of the spear, the shaft made from ash: δόρυ μείλινον (Il. 5, 666); finally, it is the spear itself, inasmuch as it is made of wood. All these are specifications of the sense “wood,” just as in French, where bois may be applied to a bed, an orchestra or a stag.

On the other hand, drū̂s did not always designate the “oak” in Greek. The ancients tell us so quite explicitly: according to the testimony of a scholiast of the Iliad (ad Il. 11, 86) δρῦν ἐκάλουν οἱ παλαιοὶ πᾶν δένδρον ‘the ancients called any tree drū̂s’. This is confirmed by the usage of writers; thus, Sophocles, Trach. 766 δρῦς πίειρα ‘the resinous tree, the pine’. The word became specialized at an early date. Already in Homer, drū̂s is the oak, the “tree” par excellence, associated with certain cults, like the prophetic oaks of Dodona. But this specialization occurred in the course of the history of Greek and at a recent period, since it did not obliterate the memory of a time when drū̂s designated “tree” in general, in accordance with the testimony of all the other languages, where the corresponding term signifies “wood, tree” and not “oak.” Further, we find in Greek itself the original sense of drū̂s in the derivative drū̂ás, which designated the mythological beings, the Dryads: these are nymphs which reside in trees, and not in oaks in particular.

There is another Greek form which is connected with drū̂s. This is déndron (δένδρον), Homeric déndreon (δένδρεον) ‘tree’, the result of a dissimilation of *derdrewon, a reduplicated form of the type called broken reduplication (cf. Lat. carcer from *karkros, Gr. karkínos).

Here, too, the sense of the root is “wood, tree.” Thus we see how all these testimonies converge and locate in a comparatively recent phase of Greek the development of the term drū̂s from the ancient sense “wood, tree,” to that of “oak.” It follows that Osthoff’s account should be exactly the reverse. The sense of “oak” is the latest phase, and one limited to Greek, of an evolution of which the intermediary step is “tree” and which may proceed from an original concept such as “to be firm, solid.” We find an exact parallel to this evolution in modern Iranian. The Persian name for “tree” diraxt, Middle Iranian draxt, is an ancient verbal adjective draxta– (the participle of drang-), the literal meaning of which is “what is steady, what is firm”; the relationship is the same as that of Greek drū̂s to *dreu-.

It can be seen that the restriction in sense which leads to “tree” and “oak” depends on local conditions. In fact the development did not take place precisely in Germanic, where *dreu remains the name for “tree” in general (Got. triu, cf. Engl. tree), while for “oak” there is a special term *aik– (German Eiche).

We are now able to reconstruct the development of Indo-European forms along different lines. From this root *dreu– come the adjectives Skt. dhruva– (the dh is secondary, of analogical origin; it replaces an ancient d), Ir. druva– ‘solid, firm, in good health’; with an initial su-, Slavic sŭdravŭ, ‘salvus, healthy’; in Baltic, Lith. drutas ‘strong, solid’ (cf. Pruss. druwis ‘faith, guarantee’, druwit ‘believe’, ‘to have faith’). In Greek (Argolic dialect) dro(w)ón is translated by iskhurón ‘strong’ according to a gloss of Hesychius. This is a development into which the whole family of Treue (Gothic triggws ‘faithful’, ‘loyalty’) naturally fits.

But on the other hand *dreu– furnishes also an adjective *drū ‘strong, resistant, hard’ which has become the word for “tree.” It follows from this that the lexical development must be placed at different levels: the sense of “fidelity,” peculiar to Germanic, is directly connected with that of the Indo-European root, whereas the sense of “tree” was an early specialization which occasionally, as in Greek, alone survives.

Here we can see in its full force the distinction between signification and designation and how great the gap between them can be, often so big that the designation gives no clue to the signification, if semantic pointers are not available. [8]

The relationships of “trust” and “fidelity” find other expressions which we shall study particularly in the Germanic languages. One of these words is used as a term of nobility and as a military term. Our study may begin with the Gothic word gadrauhtswhich in the New Testament translates στρατιώτης ‘soldier’. It is composed of a prefix ga-, indicating community, and a derivative in –ti from the verb driugan, which translates στρατεύεσθαι ‘to wage war, take the field’. From the same abstract noun drauhti– comes the denominative present drauhtinon ‘στρατεύεσθαι’ and the compound drauhtiwitoþ ‘στρατεíα, combat’, where the second element signifies “rule, law.” Outside Gothic, the abstract in Germanic takes on a different sense: Old Icelandic drōt, and the corresponding forms in other dialects, designate the “armed retinue,” the “troop”; thus Old English dryht, Old Saxon druht, Old High German truht. Especially notable is the nominal derivative of *druhti-; it furnishes in its turn a form in –no-which designates the “chief, “lord”: Old Icel. drottinn, Old Engl. dryhten, Old High Germ. truhtin. The Icelandic feminine drottning‘queen’ is still preserved in the Scandinavian languages.

Such is this Germanic word-family, the morphological relations of which are clearly apparent: an abstract noun, Goth. drauhti-, and a derived noun, literally “he who has the same drauhti-, to designate the soldier. On the other hand, another derivative in –no signifying ‘chief’ is formed on the basis of the abstract druhti-. These are the facts to be sited in the semantic context which will illuminate them.

The proper sense of these terms is recovered by comparison with a neighboring language, Slavic, and to some extent in Baltic. From this it emerges that “troop” and “chief of the troop” develops from a much more general sense, that of “friend.” In Old Slavic and in the modern Slav languages, drugŭ ‘φίλος’, ‘ἑταῖρος’ signifies “friend, companion.” The notion of a bond, of friendship, is so strong that the adjective, when repeated, may render the notion of reciprocity, “the one, the other”: Russian drugdruga. The same sense is found in Lithuanian, where draugas, with a different vocalic grade, signifies “friend, one of a couple, of a pair”; hence the abstract noun draugẽ ‘friendship, company, group of friends’. Baltic utilizes this nominal stem in a grammatical function, Lithuanian draugé ‘with’. Thus the Old Prussian compound noun draugiwaldūnen signifies ‘he who shares the inheritance, the co-heir’, German ‘Mit-erbe’.

The interest in this confrontation of German, Slavic, and Baltic is the light it throws on the proper signification of these Germanic words. We have here the notion of “company,” specified in the peculiar condition indicated in Germanic: a warrior friendship. Old Slavic preserves a parallel expression, the collective term družina ‘comrades in arms, συ-στρατιῶται’. The Gothic word for “soldier” gadrauhts, literally “he who is part of a companionship, a friendship,” understood as a collective term the group of people who are bound together by common service in war. The abstract word drauhts is “warrior companionship”; drauhtiwitoþ‘στρατεία’ is “combat” as the “rule of the *drauhti-.”

Let us now consider Old Icelandic drottinn and its group. The Germanic form *druxtinax, going back to *druktinos, is an example of a well-defined mode of formation: these are the secondary derivatives formed like Latin dominus, which designate the person at the head of a certain social group. In the Germanic languages, this type is represented by several important derivatives: Gothic þiudans (from *teutanos) ‘king, chief of the community’, kindins (from *gentinos) ‘chief of the gens’, parallel with tribūnus from tribus. In Old English dryhten ‘lord’ (in the Christian texts ‘the Lord’) represents *druktinos ‘chief of the drukti’.

This type of relationship was characteristic of ancient Germanic society. An illustration is found in Tacitus, independent of the terms we are trying to interpret and so all the more precious, in chapters XIII and XIV of the Germania. The historian describes the manner in which the Germans fight, how they assemble, how they are organized in companies, and the relations between the companies and their chief:

Noble birth or the illustrious deeds of their fathers bestow on some the rank of a prince from early childhood; the others attach themselves to chieftains, who are in the full vigor of manhood and ripe in experience; and the role of companion is nothing to be ashamed of. It even confers distinction, depending on the esteem of the prince to whose retinue a man belongs. Among these comites there exists a singular rivalry to occupy the first place beside their prince; the princes for their part vie with each other as to the most numerous and the most courageous companions.

This reminds us of the relations between the princeps and his comites: the princeps is here called ‘drottinn’ and the comitesgadraunts’. A certain correlation is established between the historian’s description and the analysis of the vocabulary.

Emperor Augustus, “First Citizen”, statue in Rome

The formation of gadrauhts is repeated in Gothic in the synonym gahlaiba ‘συ-στρατιώτης’, ‘companions in arms, comrades’, literally ‘he who shares the same bread’. It seems evident that there is a close relationship between Gothic gahlaiba and Latin companio: one of the two is a calque of the other. Probably gahlaiba is the original and companio the imitation.

The name for the “army” is a term common to the Germanic dialects: Gothic harjis, Old Icel. herr, Old High Germ. hari. It appears already in the form hari– several times in the Runic inscriptions. It is further also met with as Hario-, Chario– in the Germanic proper names which have been handed down by classical authors.

This term has a counterpart in Celtic; the form harja coincides exactly with Middle Irish cuire < *koryo ‘army’. This is confirmed by the names of Gaulish peoples: the VocoriiTricoriiPetrucorii are so named because they have two, three or four troops; thus they are constituted by a union of groups of variable numbers. Here, too, Baltic, if not Slavic, has a corresponding form: Lithuanian karias, Old Prussian karjis ‘army’.

It is possible that this comparison extends beyond the western world, if we accept the Old Persian kāra as related, a word which signifies in certain passages of the Achaemenid inscriptions “the people” and in others “the army” and so denotes “the people in arms.” In this case the correspondence is less exact. The vocalic grade is different; it has a long vowel and it is not a form in *-yo. Further, kāra-, which recurs in the Middle Persian kārčār, Persian kārzār ‘combat’ is isolated and peculiar to Persian. There is no comparable Indo-Iranian term.

We may now try to make the meaning of the term in Germanic more precise with the help of an ancient mythological designation: Old Icel. Herjan, the name or surname of the great god Odin. This name is remarkable even in its formation; it belongs to the same type of derivatives in –no– mentioned above apropos of the words for “chief.” Herjan rests on *koryonos, ‘chief of the army’. The name of Odin himself, i.e. Wotan, is also formed in this manner: *Wōdanaz ‘chief of the Wōda’, ‘of the frenzy’, or ‘the frenzied army’.

Thus in his two names the great god is designated as the chief of a group: as Odin, he is the chief of the frenzied group which perpetrate their misdeeds in his name; as Herjan, he is the chief of the troop whose mythological name is also known to us, the Einherjar, the dead warriors who inhabit Walhalla and fight under his orders. Odin in this guise is the god of the dead. This is the troop which he commands, which constitute his proper Heer ‘army’.

How do they fight? There is a correspondence between the practices of the terrestrial Heer and those of the same Heer of the next world. There is the same grouping, infernal or terrestrial, there are the same relations between the members of that group and its chief.

Here, too, Tacitus throws much light on the sense of the words in question and his text, on the other hand, is illuminated by a study of these words. In chapter XLIII of the Germania he describes the appearance which these warrior peoples assume: “Those fierce men improve on their savage nature by enlisting the help of art and time: they blacken their shields, they dye their skin, and they choose the darkest nights for battle. The horror alone and the darkness which envelops that doleful army (feralis exercitus) spreads terror: there is no enemy who can withstand that strange and, so to speak, infernal aspect; because in each battle the eyes are the first to be vanquished!” Who are such people? They are the Harii. Tacitus here describes what was later called *Wuotanes heri (German wütendes Heer), the “frenzied army” or the “army of Wotan,” disguised as the army of the dead: they take on the appearance of infernal beings (it is a masquerade) choosing the night for fighting, to strike terror into their enemies; it is an irruption of the dead among the living. Such a masquerade is supposed to represent Odin’s army in his character as Herjan, imitating on earth the exploits of Odin’s band, those which the epic calls Berserkr, literally “those who are disguised as bears.”

The Germanic name of the “army,” Gothic harjis, is defined by these conceptions and also in its lexical connections as a devastating troop: the proper activity of the Heer is characterized by the derived verb Icel. herja, Old High Germ. herian ‘to make a foray’, German heerenverheeren ‘to devastate’. In this linguistic, ethnographic, and mythological complex, we discover the structure and function of the Heer, which is something quite different from exercitus in Latin or laós in Greek. It is a grouping of the same kind as that described by Tacitus in chapters XIII and XIV of the Germania in the passage cited above to illustrate the notion of drauhti-: restricted groups devoted to a common life and a warrior companionship by loyalty to the chief whom they follow, occasionally sallying forth to plunder or to tribal combat. It is quite a different conception from the philía of the Hellenic world, which is a normal relationship between the members of large groups, whether family or tribe, sharing the same laws, speaking the same tongue and bound by ties of hospitality. In Germanic we have an exclusive friendship between man and man, in a masculine society, devoted to the practice of arms: harjisdrauhti, like German trauen, all refer to his complex of ideas and institutions.

However, is this term limited to the western European world? The Greek term koíranos (κοίρανος) ‘chief’ has often been connected with harjis, etc. It is curious, in fact, that the formation of koíranos coincides exactly with Icelandic herjan ‘chief of the army’, and this suggests that we have in Greek the same name for the army, in the form *koryo-. We must therefore define more closely the sense of koíranos which is rather vaguely translated as “chief.”

In Homer, the koíranos exercises the functions of commander, and the term, taken in this sense, provides a derivative verb koiranéō ‘to act as koíranos’. For instance, Il. 2, 207: “Thus koiranéōn, he went through the ranks of the army…”; koiranéōn(present participle) consists in reprimanding some and encouraging others; in calming down those who are excited and giving confidence to the less courageous. As for those who want to impose their views and to meddle by giving advice to their chief, he reminds them (ibid, v. 204–205): οὐκ ἀγαθόν πολυκοφανίη εἷς κοίρανος ἔστω, είς βασιλεύς … ‘polukoiraníē is not a good thing: let there be only one single koíranos, one basileús’. For the poet, different from the basileús, the koíranos is not a war lord; he never takes part in the battle himself nor is he found at the head of his troops. He goes among the ranks to make his personal authority felt. Nor does he preside over the debates in the assembly. In the Odyssey (18, 106) the beggar Iros takes it on himself to chase away those who come to beg in their turn; he provokes from Odysseus the advice not to act as a koíranos, that is to say to meddle by giving orders, by administering reprimands. So the koíranos is here again different from a fighting chieftain. In Homer, as in non-Homeric texts, koiraneîn is the activity of a local potentate exercising his authority over the people of the household rather than over the whole army. If in the Odyssey there are several passages in which the suitors koiranéousi, this is because they give orders to domestics and behave like masters. But it would seem that we cannot regard the koíranos as the military chief at the head of a given unit. The title corresponds to a very different function from that of the Nordic herjan.

Another question is the connection which there may be between koíranos and the Hittite kuirwanaš (variants kuriwanaškurewanaš) ‘independent, autonomous, not a vassal’. As far as it can be defined, the Hittite term seems only to have a fortuitous resemblance to koíranos. It is even possible, to judge by the variations in form, that it comes from a local language. It is not clear what to make of the fact that the proper name Koíranos is borne in Homer by a Lycian and a Cretan. Similarly, it is impossible to interpret in one way or another the absence of the term koíranos in Mycenean.

Part II

The expression par excellence for the notion of “loyalty,” the one which is the most general and at the same time the best characterized in western Indo-European, is the Latin fidēs with its etymological family. It is attested in several spheres of usage, i.e. with religious, moral, philosophical, and even legal senses. We shall now consider this group of words in order to define as far as we may the modalities of the notion by study of the normal relations.

To the family of Latin fidēs corresponds in Greek that of peíthomai (πείθομαι). The verbal form appears first in the middle, the present active peíthō ‘persuade’ being secondary. It was coined at a fairly late date from peíthomai ‘obey’. In accordance with an ancient morphological alternation, peíthomai has as its perfect the active form pépoitha, like gígnomai : gégona. This root provided an abstract noun pístis ‘trust, faith’, with an adjective pistós, ‘faithful’. From pistós comes a new present tense pistoûn‘to make trustworthy, to oblige, to bind by promise’ and also pisteúō ‘to have faith’, which has persisted.

Apart from Latin and Greek we can only cite with the same sense a noun form in Albanian  ‘oath’, from *bhoidā. There are numerous other phonetically comparable forms, but the sense is so different that we can not justify the relationship which the form suggests: this is where the difficulties of the problem begin. The facts are first those of Germanic: the Gothic form beidangoes back to *bheidh-, that is the same prototype as Latin fidēsfoedus, but the Gothic verb means ‘προσδοκᾶν, to expect, to await, to endure’, the same as Old Icel. biđa. Further, with another grade of the root, we have Gothic baidjan, with a different meaning again, because it translates Greek anankázein ‘compel’, just like Old Saxon bēdian ‘compel’, ‘force’. The sense of “constrain” permits however a connection with the Slavic běditi, which translates the same verb anankázein, and with the noun běda, ‘anánkē, necessity, compulsion’.

These connections are registered in all the etymological dictionaries with the uncertainties and doubts imposed by the disparity of the meanings.

We do not venture either a firm rejection or adoption of these correspondences seeing that we have no means of either justifying or refuting them. It is, however, important to know how far we can extend the comparison. Must we limit ourselves to Greek and Latin forms for the reconstruction? But if Germanic and Slavic forms are to be included, this modifies the semantic data. Before coming to a decision it will be necessary to examine the sense of the terms in those languages where it can be rigorously defined.

Let us first consider the Latin words. We must first state that the sense of fidēs is defined inaccurately in our dictionaries, so inaccurately as to make it impossible even to understand the construction of its first uses. To study it we must have recourse to the article on fidēs in the Latin Thesaurus, where the different meanings are correctly classified.

If we continue to translate fidēs with “faith,” certain essential expressions like fidem habere, fidēs est mihi, frequently met with in the language of comedy, risk being understood in exactly the opposite sense: thus Plautus, Pseudolus 467: parvam esse apud te mihi fidem ipse intellego. If we translate mihi fidēs est with “I have faith (in you), I give (you) my confidence” we arrive at exactly the opposite of what it actually means, which in fact is “(I have known for a long time that you despise me because) I understand well that you have only very little confidence in me.” Another example in Plautus, Amph. 555: facis ut tuis nulla apud te fidēs sit is to be understood in the same way: “You have no confidence in your people.”

The context and the authentic syntax of this turn of phrase impose a translation which seems to reverse the expected connections: fidēs est mihi apud aliquem signifies “somebody has confidence in me.” To translate fidēs more literally, let us replace “confidence” with “credit.” The literal translation of fidēs est mihi apud aliquem becomes “I have credit with somebody”; this is really the equivalent of “I inspire confidence in him” or “he has confidence in me.” Thus the Latin notion of fidēs establishes between the partners an inverse relationship to that which we generally understand under the notion of “confidence.” In the expression “I have confidence in somebody,” the confidence is something belonging to me which I can put into his hands and which he disposes of. In the Latin expression mihi est fidēs apud aliquem it is the other who puts his trust in me and it is at my disposal.

Thus the term fidēs is bound up with the construction est mihi, the proper expression of possession; and this “possession” is determined by the preposition apud ‘chez’, indicating the partner. The “possessor” of the fidēs thus holds a security which he deposits “with” (apud) somebody: this shows that fidēs is really the “credit” which one enjoys with one’s partner. All the early examples confirm this.

This term figures in still another well-known turn of phrase where the sense also requires rectification. This is the appeal: pro divom fidem made to obtain the help of the gods, or again: di, obsecro vestram fidem, ‘O gods, I beseech you for your fidēs’. Since fidēs designates the confidence which the speaker inspires in his interlocutor, and which he enjoys with him, it follows that it is for him a “guarantee” to which he can have recourse. The fidēs that mortals have with the gods assures them in return of a guarantee: it is this divine guarantee which the speaker invokes in his distress.

Once we have penetrated into these syntactical and semantic relations, it is the French phrase avoir confiance en quelqu’un ‘to have confidence in someone’ which looks peculiar. It is right to say “je donne ma foi, j’accorde ma confiance,” ‘I give my trust, I bestow my confidence’. Something of mine is in effect given to somebody who now possesses it (“he possesses my confidence”). But how to explain that we also say “to have confidence” in somebody? How can one give a thing and have it at the same time? The answer should not be sought in French or English itself; the expression “avoir confiance” ‘to have confidence’ is incomprehensible except as a translation of the Latin fidem habere. We must thus explain fidēs in this new construction which is quite different from the other. This time it is the verb which we must consider. In fact, the turn of phrase fidem habere alicui is to be understood in the same manner as honorem habere alicui ‘to bestow honor on somebody’, and signifies thus “to bestow on somebody the fidēs which belongs to him.” Thus Terence, Eun. 197: forsitan hic mihi parvam habeat fidem ‘perhaps this man will have little confidence, will bestow on me slight fidēs’.

Here we see the relation between hic mihi fidem habet and the ancient est mihi fidēs apud ilium. By a natural development we pass in the language of rhetoric to the expression fidem facere orationi ‘to create fidēs in an oration’, that is credibility. From now on it is the utterance which possesses a fidēs and it is possible to say est orationi fidēs apud auditorem ‘the speech possesses this fidēs vis-à-vis the hearer’ and thus becomes capable of persuading him. From this by abbreviation we get fidem auditori facere, literally “to make credibility for the hearer.”

It is from this that fidēs develops into a subjective notion, no longer the confidence which is inspired in somebody, but the trust which is placed in somebody. This conversion was the essential stage in the evolution. It would be possible to follow the development of the notion in familiar phrases: se in fidem ac dicionem populi Romani tradere ‘to deliver oneself into the fidēsand power of the Roman people’; fidēs is joined to dicio, the power to dispose of somebody; or se in fidem et potestatem alicuius tradere, ‘to surrender oneself into the fidēs and power of someone’. Just like potestās and diciōfidēs is a quality acknowledged in the victor.

These equivalents bring to light another aspect of fidēs. If we review the different words associated with fidēs and the circumstances in which they are employed, it will be seen that the partners in “trust” are not in the same situation; the one who holds the fidēs placed in him by a man has this man at his mercy. This is why fidēs becomes almost synonymous with diciō and potestās. In their primitive form these relations involved a certain reciprocity: placing one’s fidēs in somebody secured in return his guarantee and his support. But this very fact underlines the inequality of the conditions. It is authority which is exercised at the same time as protection for somebody who submits to it, an exchange for, and to the extent of, his submission. This relationship implies the power of constraint on one side and obedience on the other. It is seen very clearly in the precise signification of the Latin word foedus (from *bhoides-), a “pact” established originally between two unequal partners. This is shown in certain poetic usages: omnes foedere naturae certo discrimina servant ‘all, in conformity with the laws fixed by nature, preserve the characteristics which differentiate them’ (Lucretius V, 923); has leges alternaque foedera certis imposuit natura locis ‘nature has imposed these laws and eternal pacts on certain localities’ (Vergil, Georgics I, 60). The constraining power of foedus was later extended to both parties.

The Latin forms illuminate the various aspects of the sense thanks to the phraseology of the religious and legal language. Outside Latin, these notions have become laicized and specialized. Nevertheless, the verb peíthomai in Greek “I let myself be persuaded, I obey” still enables us to recognize that “persuasion” is equivalent to, or develops to, the sense “obedience” and presupposes a constraint although the institutional form of this submission is no longer apparent.

We may now return to, and make more precise, the etymological relationships with the Germanic and Slavic forms. Up to now etymologists have left open the question whether the sense of Gothic beidan ‘to wait, bide’ should or should not be connected with that of fidēs, etc. The same is true of Old Slavic běda ‘constraint, anánkē’. Similar problems often arise if we take too summary a view of the relationships of sense. The first condition is to observe and to define exactly the terms in question in the language itself. If we examine how Gothic employs beidan ‘to expect, prosdékhesthaiprosdokân’, it will be noticed, particularly in Luke II, 25 “he was a just and pious man” beidands laþonais Israelis, προσδεχόμενος παράκλησιν τοũ Ἰσραήλ, ‘who expected the consolation of Israel’. Here the “expectance” is a “confidence” in the fulfilment of the prophesy of Isaiah (33, 20). Mark XV, 43 was silba beidands þiudangardjos gudis (Joseph of Arimathea, a notable member of the Council) ‘who also expected the kingdom of God’. Here, also, “expect” is equivalent to “place one’s confidence in…” Luke II, 38 þaim usbeidandam laþon Jairusaulwmos ‘to those who expected the deliverance of Jerusalem’; it is still an event expected with confidence that is given by conviction. This is indirectly confirmed in the context of I Cor. XIII, 7 where gabeidiþ ‘ὑπομένει, endures’ follows þulaiþ‘excuses’, galaubeiþ ‘believes’, weneiþ ‘hopes’. There thus is in Gothic no break with the ancient sense of *bheidh-, but only an evolution from “put one’s confidence in somebody or something” to “expect,” and even if it is taken in an ordinary sense, this verb always refers to a hopeful expectation.

Nor is there any difficulty in admitting that beidan has its causative in baidjan. Here, again, scholars have found an insurmountable obstacle in the sense of baidjan, which translates Gr. anankázein ‘constrain’; how could “constrain” be the causation of “expect”? The fact is that the following has not been taken into consideration: Gothic uses two different verbs to render anankázein. One is nauþjan ‘to exercise a physical constraint’ and the other baidjan, indicating only a moral constraint, which is that of persuasion (cf. II Cor. XII, II; Gal. II, 3, 14). It is thus possible to imagine that the connection between beidan and baidjan is analogous to that of Gr. peíthomai ‘to trust somebody’ and peíthō ‘to get somebody to obey’. The same is true of Old Slavic běda ‘constraint’. In this light the old unity can be restored and we can see that, as between the senses of the Greek and Latin forms and those of Germanic and Slavic, there was a weakening and especially a loss of the institutional sense. This is in the main due presumably to the emergence of another expression for faith and fidelity in Germanic, i.e. Treue and the related terms.

The history of fidēs goes beyond its etymological relatives. It has long been noticed that fidēs in Latin is the abstract noun corresponding to a different verb: crēdō. This suppletive relationship has been studied by A. Meillet [9] who has shown that the ancient connection between crēdō and fidēs was revived in Christianity: it was then that fidēs, a profane expression, evolved towards the sense of “religious faith” and crēdere ‘believe’ towards that of “to confess one’s faith.”

We must here anticipate the conclusions of an analysis which will be found below (Book One, Chapter 15) in order to demonstrate what predetermined to some extent that fidēs and crēdō should function in this suppletive way. Crēdō, we shall see, is literally “to place one’s *kred,” that is “magical powers,” in a person from whom one expects protection thanks to “believing” in him. Now it seems to us that fidēs, in its original sense of “credit, credibility,” implying dependence on the one who fidem habet alicui, designates a notion very close to that of *kred. It is easy to see, once the old root noun *kred was lost in Latin, how fidēs could take its place as a substantive corresponding to crēdō. In these two terms we are back once again with notions in which there is no distinction between law and religion: the whole of ancient law is only a special domain regulated by practices and rules which are still pervaded by mysticism.


  1. On timḗ and its group cf. Book 4, Chapter 5.
  2. For a detailed analysis of the “gift” vocabulary see our article “Don et échange dans le vocabulaire indo-européen,” L’Année Sociologique, 3rd Series, vol. II, 1951, pp. 7–20.
  3. Cf. our article “Don et échange dans le vocabulaire indo-européen,” already cited.
  4. For the meaning of némō we may refer to our analysis of némesis in Noms d’agent et noms d’action en indo-européen, Paris, 1948, p. 79.
  5. On Gr. pérnēmi and Lat. emo, see Book One, Chapter Ten.
  6. For the semantic study of pot(i)-, reference may be made to our article “Problèmes sémantiques de la reconstruction,” Word X, Nos. 2–3, 1954, and Problèmes de linguistique générale, Gallimard 1966, pp. 301ff.
  7. On the root mei– see our article “Don et échange…” quoted above.
  8. For *doru-/*dreu– see our article “Problèmes sémantiques de la reconstruction” already cited.
  9. Mémoires de la Société de Linguistique de Paris, XXII, 1922, 215ff.

Originally published by The Center for Hellenic Studies under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported license. From “Indo-European Language and Society”, by Émile Benveniste (University of Miami Press, 1973).