Chapter 16: Lending, Borrowing, and Debt

In contrast to Bartholomae, who distinguishes two roots par-, it is shown that the Iranian derivatives (and the Armenian ones) of par-, from which comes Iranian *pr̥tu-, and from it Armenian partk^c ‘debt’, can be attached to a single basic meaning “compensate by something levied on oneself, on one’s own person or one’s own possessions.” Lat. par ‘equal’ can be brought together with par-in Iranian.

In Latin, debere ‘to owe’ does not imply the receipt of something from someone to which it must be given. The technical expression pecunia mutua, on the contrary, designates precisely the twofold movement, i.e. going, coming back of the same sum of money, without any interest.

In Germanic, the specialization of leihv- < Indo-European *leik ^w- (cf. Gr. leípō ‘to leave’) in the sense of “lend” depends both on the notion of “vacating” attached to this root and on the existence of another verb—letan—for “leave.” On the other hand, to designate “debt,” Gothic, which has a verb for “to be obliged to” (in general), has had to borrow another term from Celtic.

Again in Gothic, the vocabulary of “lending,” which apparently was not very precise, in fact comprises two different notions—one is of long standing, that of a loan as a personal transaction, the other is recent, that of a loan on interest as a professional activity. Analogous facts can be observed in Greek.

The purpose of this chapter is to show how, independently in several languages, in Iranian, Latin, Gothic and Greek, the technical terms relating to “debt,” “loan” and “borrowing” were constituted by specialization and differentiation of more general terms or those belonging to a different order of ideas. We shall encounter, however, apart from special terms which are the product of an evolution peculiar to each language, on the one hand a term of considerable generality and on the other morphological processes common to the group of words connected with these notions.

In the eastern Indo-European region, there is a series of Iranian forms without (in the present state of research) sure correspondences elsewhere, which are difficult to differentiate in Iranian itself. These are the derivatives from the Avestan root par-.

The distinctions between the words which derive from it are not clearly made in the authoritative dictionary, that of Bartholomae. The first task must, therefore, be to attempt an analysis which will enable us to regroup words dispersed in several articles. Bartholomae in fact distinguishes two roots (1) par- ‘to pay back equal amounts’, (2) par- ‘to condemn’. In my opinion we must bring together the forms deriving from both roots to make up a single family: these forms are partially congruent in the two articles in Bartholomae’s dictionary. They are generally used in the passive: e.g. pairyete, the present tense common to both roots par-: “to be compensated” or “to be condemned.”

An example will show the context in which these forms appear. The derivative ā-pərəti, with the preverb ā and the suffix -ti, occurs in the company of a middle participle pārəmna- from the same root in the following passage: “Such is the čiθā, such is the āpərəti for the faithful who repents (pārəmnāi)” (Vidēvdāt 8, 107).

The abstract noun āpərəti is accompanied by čiθā ‘expiation, compensation’, the two together indicate a reparation made to expiate a sin against religion. Āpərəti is also found as equivalent to yaoždāθra-, an action to make somebody or something ritually appropriate which is polluted and hence unsuitable for religious use.

Two other derivatives are used especially in the code of purity called Vidēvdāt: pərəθā- ‘corporal punishment’, ‘fine’, something that is given to atone for a sin; and the negative adjective anāpərəθa- ‘not to be compensated’, ‘inexpiable’ applied to šyaoθna- ‘action’.

We next have a series of forms which have been linked to another root par-, but which actually ought not to be dissociated from those just discussed. They are legal expressions frequently found in the Vidēvdāt: from the neuter pərəθa ‘expiation, compensation’ (which is implied by the adjective anāpərəθa which we have just mentioned) certain compounds were made: tanu-pərəθa, pərətō-tanu, pəšō-tanu (the last two forms are merely orthographic variants), the literal meaning of which is “of whom the body (tanu) is condemned, serves as a compensation,” an adjective qualifying those who have committed certain crimes. Very characteristic, too, is the conjunction in one and the same expression of the compounds dərəzānō-pərəθa—‘he whose compensation is heavy’—with the noun pāra- ‘debt’. The Avestan vocabulary enables us to discern a set of ideas which pertain both to religion insofar as they are connected with “expiation” or “compensation,” and to economic relations. This is confirmed by the indirect testimony of Armenian, which has borrowed at all periods of its development a considerable number of Iranian words. Given the large gaps in our knowledge of Iranian for certain periods, Armenian helps us to reconstitute lexical families which are defective or insufficiently represented in Iranian.

Such is the case here. We have in Armenian partk ^c ‘debt’ (with k ^c of the plural which is normal in abstract words), genitive partuc ^c, a stem in -u, which is otherwise unknown in Iranian. We have thus an opposition of two abstract formations : ā-pərəti and *-pr̥tu, that is to say the two forms in -ti and -tu respectively. In Armenian partk ^c ‘debt’ designates also “obligation” in general, the fact of “owing,” just like German Schuld. Hence such expressions as part ē inj, literally “there is a debt, a duty for me,” i.e. “I owe, I have an obligation to” (negative

^c ē part inj ‘I need not’), whether it concerns a moral obligation or a debt. With the common suffix -akan, the adjective partakan ‘debtor’ has been derived from par-, which may be construed as a predicate, partakan ē. Later the word became specialized also in compounds of which both components are Iranian in formation: partavor ‘he who bears a debt or an obligation; liable’; and in particular part-a-pan ‘debtor’, literally “he who preserves a debt.” From partapan was created the opposite term partatēr (in which tēr is an Armenian word for “master”), literally “master of the debt,” that is “creditor.” From this comes a number of new derivatives: first the verb partim ‘I owe, I am obliged to’; then a technical term which may be taken from Iranian, the compound part-bašxi, the use of which explains the formation. One says in Armenian “to give one’s own fortune as a partbašxi for others,” which means “to settle the debts of others.” This compound *pr̥tu-baxšya- (this being the original Iranian form of the Armenian loanword) will have meant “the settlement of a debt”: this is a technical expression of the legal language.

We thus have at our disposal a fairly considerable collection of forms. We must now pay closer attention to the characteristic suffixes of these terms. The word for “debt,” *pr̥tu, is to be defined literally as “a thing to compensate,” hence “obligation” in general. This interpretation is suggested by the suffix -tu which implies an aptitude or eventuality. On the contrary, with the suffix -ti, the Av. derivative āpərəti represents the expected sense of “effective compensation,” hence (and this is the attested sense) “debt effectively settled,” which is different from *pr̥tu—“debt” that is still to be settled.

The notion of par- in Iranian is much wider than our notion of “debt”: it is everything which is owed by way of reparation, by one who is guilty of an offense. Thus there is after all only one root par- ‘to compensate by something levied upon oneself, one’s own person or property’; this meaning accounts for the whole lexical family just reviewed.

We find a correspondence outside Iranian (the root is unknown in Indic as far as I am aware): this is the Latin adjective par, paris, indicating parity or equality. There is no primary verbal root in Latin: paro, comparo are derivatives of the adjective par. In Umbrian, too, pars (Lat. par) is only a noun.

The sense permits the equation: it is one of those survivals which connect Latin with the eastern group of the Indo-European languages, and the correspondence is all the more instructive because it supplies the starting point of the technical development which took place only in Iranian and produced the term for “debt.” It is largely from religious notions that these legal expressions have been constituted.

We must be careful to distinguish homophonies. The group of Latin and Iranian forms has nothing to do with those which were studied apropos of the notion of “sell,” which derive from a root of the same form: perdō, epérasa, pipráskō. As we have seen, the expression for “sell” goes back in Greek itself to the sense “to transfer, to take abroad.”

This is far from the sense “to compensate,” and the two roots *per- have nothing in common, either in their sense or their dialect distribution.

The sense of Latin dēbeō ‘owe’ seems to result from the composition of the term dē + habeō, a compound which is not open to doubt since the Latin archaic perfect is still dēhibui (for instance in Plautus). What does dēbeō mean? The current interpretation is “to have something (which one keeps) from somebody”: this is very simple, perhaps too much so, because a difficulty presents itself immediately: the construction with the dative is inexplicable, debere aliquid alicui.

In Latin, contrary to what it might seem, debere does not constitute the proper expression for “to owe” in the sense “to have a debt.” The technical and legal designation of the “debt” is aes alienum in the expressions “to have debts, to settle a debt, in prison for debt.” Debere in the sense “to have debts” is rare; it is only a derived usage.

The sense of debere is different, although it is also translated by “to owe.” One can “owe” something without having borrowed it: for instance, one “owes” rent for a house, although this does not involve the return of a sum borrowed. Because of its formation and construction, debeo should be interpreted according to the value which pertains to the prefix de, to wit: “taken, withdrawn from”; hence “to hold (habere) something which has been taken from (de) somebody.”

This literal interpretation corresponds to an actual use: debeo is used in circumstances in which one has to give back something belonging to another and which one keeps without having literally “borrowed” it; debere is to detain something taken from the belongings or rights of others. Debere is used, for instance, for “to owe the troops their pay” in speaking of a chief, or the provisioning of a town with corn. The obligation to give results from the fact that one holds what belongs to another. That is why debeo in the early period is not the proper term for “debt.”

On the other hand, there is a close relation between “debt,” “loan” and “borrowing,” which is called mutua pecunia: mutuam pecuniam solvere ‘pay a debt’. The adjective mutuus defines the relation which characterizes the loan. It has a clear formation and etymology. Although the verb muto has not taken on this technical sense, the connection with mutuus is certain. We may also cite munus and so link up with an extensive family of Indo-European words which, with various suffixes, denote the notion of “reciprocity” (see above, Book One, Chapter Seven). The adjective mutuus indicates either “loan” or “borrowing,” according to the way in which the expression is qualified. It always has to do with money (pecunia) paid back exactly in the amount that was received. Lending and borrowing are two aspects of the same transaction as the advance and repayment of a given sum, without interest. For a loan at interest there is another word, fenus.

The relation between the sense of muto, which is translated “to change,” and mutuus is mediated by the notion of “exchange.” Muto means “to change” something (a garment, for instance) for something equivalent. It is a substitution: instead of the thing given or “left,” something identical is received. The meaning remains the same whatever noun appears as the object of the verb: mutare vestem, patriam, regionem, means to “replace a piece of clothing, a country, a region by another.” Similarly, mutuus qualifies what is to be replaced by an equivalent. There is an evident link with munus which, although bound up with a different set of ideas, is connected with the same kind of notion. The root is Indo-European *mei-, denoting exchange, which has produced in Indo-Iranian Mitra, the name of a god, besides meaning “contract.” We have studied above the Avestan adjective miθwara, Skt. mithuna, exhibiting the same radical suffix -t- as mūtuus. The sense is “reciprocal, making a pair, constituting an exchange.”

But the sense of munus, which is particularly complex, developed in two groups of terms which we had occasion to study above and which denote both “gratuity” and “official duty or function.” Such notions are always of a reciprocal character, implying a favor received and the obligation to reciprocate. This explains both the sense of “administrative duty,” “official function,” and that of “a favor shown to somebody,” because what is concerned is “public service,” that is to say, an office conferred on somebody who honors it by keeping it within limits. The “favor” and the “obligation” thus find their essential unity.

We shall now consider the same notions in the Germanic languages. The expressions are entirely different: Got. leihvan ‘lend’, Old High German līhan, Old Icel. lān; modern English loan, German leihen, etc. The meaning is constant and well established from ancient Germanic onwards; an indirect proof is that these terms have passed into Slavic. OSl. lixva translates Gr. tókos ‘interest on money, price’, and the word is pan-Slavic. These words belong to the family of Greek leípō (λείπω), Lat. linquo ‘to leave’. The early specialization of this verb, the sense “to leave” of which is general in Indo-European, poses a problem. We must try and determine the conditions in which this specialization (which is not general) took place. Thus in Indo-Iranian, rik- and in Armenian lk ^c anem, a nasal present stem, mean only “to leave” or “to remain.” This curious development of sense was studied by Meillet, [1] who stressed the fact that it is not sufficient to explain “lend” as “to leave something to somebody.” The problem is precisely to find out how the term has become restricted and specialized.

Meillet observed that we have in Indo-Iranian from the same root *rik- the Sanskrit derivatives reknas and Avestan raēx-nah-, both denoting “inheritance,” and they correspond exactly. These noun forms in Indo-Iranian, characterized by the suffix -nes, recall the Germanic noun forms, like lehan. It was because of the sense “loan” acquired by lehan that the Germanic verb became specialized in its turn in the sense of “to lend.”

This root *leik ^w- which is translated by “leave” or “remain” according to whether the verb has an object or not in fact signifies: “to be in a deficient state,” “to be wanting, absent,” “to be missing from the environment where one ought to be.” The Homeric perfect tense léloipa does not mean “I have left” like reliqui, the transitive perfect, but “I am in a state of deficiency,” an intransitive perfect which in spite of its construction could be active: leloipó̄s signifies “who is missing.” The usual definition conforms too much to the sense of the Greek and Latin terms. Skt. rik- signifies “to be missing, empty, deprived”; with the verbal adjective we have the compound rikta-pāṇi, rikta-hasta (to present oneself before somebody) ‘with empty hands’. We also note the phrase riktī kr̥ (cf. Lat. multi, lucri facio) ‘to empty’, ‘leave’, and the adjective reku- ‘empty, deserted’.

These facts are confirmed by Avestan, which offers expressions of a similar sense: a present causative in -aya-: raēčaya- ‘to make to evacuate’, literally “to make (the water) withdraw.” The sense of rik- thus will be “to evacuate, to leave something empty, of one’s presence,” but not “to remain.” The derived noun reknas designates “heritage,” not as something which one “leaves” in general, but a property evacuated, left vacant (by the disappearance of its owner).

Meillet rightly stressed the formation in -nas, which characterizes mūnus itself and a small group of words connected with property, like Skt. apnas ‘goods, fortune’, where the ap- is to be compared perhaps to ops in Latin; derived from another root, Skt. draviṇas has the same sense: “movable goods, fortune.” Here is the right place to cite Lat. fēnus ‘loan at interest’, the fē- of which evidently belongs to the group of fēcundus, fēlix, fēmina, words with a very different meaning, but which have in common the root fē- that corresponds to Gr. θη-, the original sense of which is “fecundity, prosperity.” Thus fenus evokes the same image as Gr. tókos: the interest is, as it were, the offspring of the money. We may also establish the supplementary condition which allowed this specialization. For “leave” Gothic has letan (‘to let’, German lassen) with a large variety of uses: to leave an orphan, to let somebody depart, to leave money. From this, given this range of meaning, leihv- was available for use in a special sense.

We also have in Vedic the germ of a specialized use: rik- ‘to retire from, to abandon something’, is sometimes constructed with an object noun in the accusative and an instrumental, in the meaning “to abandon the possession of something for a certain price,” and in consequence “to part with for money,” “to sell.” Certainly this is not “to lend,” but it can be seen that that rik- could refer to certain transactions.

The expression for “to borrow” and “to lend” in Germanic is a verb represented by the English borrow, German borgen, and the corresponding forms in the Germanic languages. It is a present denominative from borg, meaning “surety, guarantee”—in an ablaut relationship with the Gothic verb bairgan ‘to guard, to preserve’. The transition can be seen in Old Saxon borgjan meaning “protect,” then “to be a guarantor,” hence “lend” and correlatively “to give a guarantee,” hence “to borrow.”

The parallelism “lend/borrow” is easily apprehended in Germanic because the same verb borgen expresses the two notions. Even in Gothic, where there are separate terms, the connection is obvious: “lend” is expressed by “leave” and “borrow” by “keep,” “guard.” The lexical distinction can be dispensed with; for instance, emprunter in Old French was used for “to lend” and for “to be made to lend.”

This relationship is also observed in the Greek technical term dános (δάνος) ‘money lent at interest’ (another derivation with the suffix -nes), whence the present tense daneízō ‘to lend’. By varying the voice between active and middle this verb suffices to express both “lend” and “borrow.” However, there is as yet no satisfactory etymology for dános. If we can accept the gloss δάνας· μερίδας, the ancient sense was “part”: we must then regard dános as a derivative in -nes (neuter) form of the root datéomai ‘to share out’, comparable to the Skt. verbal adjective dina ‘shared out’. The difficulty is to explain how “to share out” could evolve to the sense of “lend, borrow.” The explanation will offer itself in a different connection later on.

For “to owe, be obliged to” Gothic has a verb skulan in a general or specialized sense, either as a material or a moral obligation. It translates both opheílō in the sense of “being a debtor” and the same verb opheílō when it serves to express in the Greek of the Gospels “to have a duty, to impose a moral rule on oneself”; skulan is also used to render méllō, which is one of the ways of expressing the future tense “I ought” with the infinitive. The perfect participle skulds, when used with “to be,” forms a periphrastic expression with an active infinitive to render the notion of obligation in the passive voice, because there is no infinitive of the passive voice in Gothic. It was therefore necessary to construct the infinitive with the passive voice of the auxiliary verb, “he ought to be called” is literally expressed as “he is obliged to call.” There is also an impersonal use with the neuter: skuld ist, which translates éxesti, deî ‘it is possible, it is necessary’.

The noun skula ‘debtor’ is construed either with a noun form or with the infinitive. It designates the one who “owes” money, is liable to some obligation, possibly some punishment, from which comes: culpable or accused in a criminal matter, etc. (cf. German schuldig ‘guilty’). In the case of a monetary debt, we have a special expression: dulgis skulans, which translates the plural khreopheilétai (χρεοφειλέται). Thus in Luke VII, 41 twai dulgis skulans wesun dulgahaitjin sumamma: δύο χρεοφειλέται ἦσαν δανειστῇ τινι, literally “there were two debtors to one creditor.” To express “those who owe a debt” the nominal derivative of skulan did not suffice; the notion had to be determined by dulgis. Furthermore, the antithetic term “creditor” is formed by means of a compound: dulgahaitja, which contains the same determinant. Thus the noun dulgs, signifying “debt,” is etymologically independent of the verb skulan ‘to owe’. This dulgs also enters into a compound which renders Gr. daneistḗs ‘he who lends’.

The remarkable fact is that dulgs is not of Germanic origin: it is a borrowing from Celtic. The Celtic form is related to a group of important terms in Irish, dliged ‘the law, the right which one has over somebody’ and the verb dligim ‘to have legally, to have the right over somebody, over something’. The verb can be constructed in two ways according to whether the subject is active or passive: in the passive, Old Irish dlegair domsa ‘right, possibility of a claim against me’; or dligim nī duit ‘I have a claim, a right over something of yours’, you owe me something, I am in the position of asserting a claim on you.

The Gothic expression dulgis skulan is doubly significant. By itself skulan and its derivatives could not specify money debt; in order to specify this it was necessary to borrow the word for “debt” from Irish. It seems, then, that the Gothic vocabulary was not sufficiently evolved to express the notions of money, loan, and borrowing in their legal context.

But the problem is still more complex. We shall try to see by direct analysis of an important text how the Gothic translator managed in a particular case. This is the parable of the pounds, Luke XIX 12-26. Faced with the constantly recurring Greek term mnâ (‘pound’, ‘mina’), Gothic seems to use several equivalents which appear to be used somewhat haphazardly. A man departs for a far-off country and entrusts ten pounds to ten servants for them to invest.

Luke XIX, 13: “he gave them ten pounds (mnâ)—taihun dailos— and he said to them: trade with (in Greek pragmateúein ‘to carry on a financial operation’) this money.” Gothic uses the imperative kaupoþ (German kaufen) ‘buy’, also “trade in money.” There is no other expression in Gothic for commerce and speculation than kaupon, formed from the Latin loanword caupo.

In verse 15, after his return, the man calls his servants “to whom he entrusted his money” until he should return: οἷς ἔδωκε τὸ ἀργύριον: “silver,” argúrion, is translated by silubr.

In verse 16, “the first man presented himself: ‘Lord, your pound has brought in ten pounds,’” skatts þeins gawaurhta taihun skattans; this time skatts takes the place of dailos for “pound.”

Similarly, in verse 18, “the second came and said ‘Lord, your pound has brought in five pounds.’” Again we see skatts and the accusative plural skattans.

In verse 20, the last man said to him: “here is your pound which I have kept tied up in a napkin”; here, again, skatts.

In verse 23, the master retorts: “why did you not put my money into a bank?” Gothic translates money by silubr (as earlier on) and the bank (Gr. trápeza) by the expression skattja ‘changer’, the agent noun derived from skatts.

In verse 24, the master continues, addressing those present: “take the pound from this man and give it to him who has the ten pounds.”

Here, pound is translated by skatt; but the ten pounds by taihun dailos. When the number changes from singular to plural, the term also changes.

In verse 25, the others protest: “Lord, he already has ten pounds” habaiþ taihun dailos.

Thus, according to the context, Gothic uses one word for “money”: silubr, but two for “pound”: skatts and daila. Furthermore, Gothic possesses, to render “silver,” substance (argúrion) or money (khrḗmata), also the term faihu (cf. above, Book One, Chapter Four). We can see, therefore, four possibilities:

What is the cause of this strange variety in a field where it would appear that Gothic had no very developed vocabulary?

Let us first consider the words for silver: silubr is a foreign word, the origin of which it is impossible to trace. It is limited to Indo-European of the north and north-east: Germanic, Baltic, Slavic. The Baltic forms are not homogeneous: OPruss. siraplis, Lith. sidãbras, Lett. sidrabs, as against OSl. srebro. The forms in these languages do not correspond exactly. The variations are such and they are so irregular that they suggest borrowing from a common source unknown to us.

The word probably denotes the material rather than the coined money. In the other Indo-European languages the term for “silver” is a designation of very great antiquity, signifying “white, brilliant,” as is witnessed by argúrion and its related terms. Gr. argúrion ‘silver’ denotes the metal as well as money. In Gothic itself, faihu is the correspondent of pecus; it does not signify “livestock,” but “wealth,” in particular “money”: philárguros ‘greedy for money or avaricious’ is translated by faihu-friks ‘desirous of faihu’,” cf. faihu-gairns ‘he who loves money’, faihu-gawaurki ‘money revenue’, the second component of which links up with gawaurkjan ‘to produce by work’, the preterite of which, gawaurhta, occurred above (Luke XIX, 16).

We thus have two terms used for mnâ. One, skatts (German Schatz ‘treasure’), has no correspondent outside Germanic. It translates mnâ ‘mina’, as well as dēnárion (δηνάριον) in spite of the considerable difference in value between the two currencies. Further, more generally, it translates argúria, argúrion ‘money’. But what emerges from this variety of terms is that skatts does not presuppose any precise definition of money; it translates different monetary values. From skatts is derived the masculine noun skattja ‘money changer’. This is the word which was chosen by the Gothic translator to render trápeza ‘bank’.

The second word, daila, is quite different: This is the only passage where it appears in this sense which, evidently, must have been usual. It belongs to common Germanic. Besides daila or dails (German Teil ‘part’) Gothic has dailjan ‘divide’ with the preverbs af-, dis-, ga-, the sense being specified by these preverbs; distribute, divide, share out. In another passage, daila translates Gr. metokhḗ ‘participation’ but, in the present series of examples, mnâ.

The master divides ten minae (dailos) among his servants. Then, one mina produces ten minae (skatts). Finally, he takes away the mina (skatts) to give to the one who has ten minae (once again dailos); the two terms seem to be used concurrently.

The contrast is a deliberate one: daila, which elsewhere is equivalent to metokhḗ ‘participation’, is here the “part” of the total sum which was evenly divided at the beginning of the story: it is also the “share” of the same sum which was given back at the end by the clever speculator. But skatts denotes the monetary unit itself, with its proper value. This fact dictates the choice: on the one hand the monetary symbol, counted in distinct units; on the other hand, the “part,” whether it results from a division, or is something which has been increased by investment. Such considerations seem to be responsible for the choice of the terms at his disposal which the translator made.

We must here take up again an analysis left in suspense. The Gothic—and Germanic—verb for “lend” is Gothic leihvan, German leihen, Engl. loan, from the root of Gr. leípō, Lat. linquo. Strange to say, the verb assumes in Germanic the sense of “to lend,” whereas everywhere else it signifies “to leave” or “to remain.”

How has the general notion of “leave” become the expression for “to lend”? Here we must expound two facts which are interconnected and serve to explain each other.

According to the testimony of Tacitus: (apud Germanos) fenus agitare et in usuras extendere ignotum (Germania, 26): “(the Germans) were not acquainted with loans at interest.” Certainly, Tacitus draws an idealized picture of Germany, but he has clearly not invented this particular feature: the Germans did not know the fenus, the loan at interest. Generally speaking, the notion “to lend” is expressed in Gothic in two ways:

1) One “leaves” to somebody the use of something belonging to one; this is leihvan, which is applied to any object whatsoever (Matthew 5, 42; Luke 6, 34-35) except money: herein lies the difference.

2) A loan of money consists of entrusting money on the condition that it yields. This notion may not be very old: Gothic, having no ready-made term, coined kaupjan ‘to speculate’.

Apparently in this society one did not lend money: only professionals practiced lending.

Retrospectively, another fact may shed some light: Gr. dános, a technical term for “loan at interest,” whence comes daneízō ‘to lend at interest’, daneízomai ‘to borrow’, daneistḗs ‘debtor’. We have mentioned above the etymological connection of dános with daíō, datéomai ‘to divide’. The Greek term is glossed méros ‘part’; dános is a neuter in -nes of the type fenus, pignus, which belongs to the vocabulary of social transactions.

But how can we link “loan at interest” with “divide”? In Greek there may be some connection, as in Gothic with daila, dails, which translates méros, merís, metokhḗ, etc. In dános we have the designation for the “participation” or “part” which accrues to professionals from their operations in money changing or lending.

Thus the notion of “loan at interest, credit, debt,” gives rise in Gothic to two different categories of terms, according to whether it concerns a professional activity or a personal transaction. Hence such different expressions as dulgis skulan and daila.

In Greek, too, we have a general verb like opheílō either for a monetary debt or a moral obligation. But where money is concerned, special terms are coined, these being derivatives from khrḗ: khrḗmata, cf. khreopheilétēs, or a term like tókos, interest in the proper sense. On the other hand, dános, daneízō denote solely the loan at interest in the varieties noted above.

It remains to consider one more verb which, originating in Latin, passed into French. This is the Latin praestare; the exact sense of the verb, in view of the range of its use, remains to be defined. Along with praestare, the adverb praesto (esse alicui) suggests a relation which finally evolved to that of French prêter ‘lend’. But we must first make clear the links between the varied uses of praestare. There are two present forms praesto in Latin: one is prae-sto ‘to keep oneself ahead, to be at the head of, to distinguish oneself’ etc., this being one of the compounds of sto. The other is the one we are studying.

Whatever the etymology of the adverb praesto, praestare must be regarded as a derivative of it. It is a present tense based on an adverb, a curious formation. In this morphological character we find the point of departure for the sense and at the same time the reason why there are so many different constructions with this verb.

The adverb praesto has this peculiarity that it enters only into a predicative and intransitive construction: praesto esse ‘to be at the disposal of, to present oneself (to view, for service)’. The problem was to convert it into the predicate of a transitive construction and to transform praesto esse into a *praesto facere. Instead of this *praesto facere, Latin coined a present derivative praestare, which has this function and thus signifies “to make something ready for,” “to put at the disposition of.” But according to the nature of the object noun, it can take on various meanings: aliquid alicui praestare may mean “to bring it about that somebody can count on something,” hence “act as guarantor, be responsible for”: emptori damnum praestare ‘to be responsible for a loss vis-à-vis the buyer’. When the object is a personal quality, the verb means literally “to make a quality apparent (to view, for the service of somebody),” hence “to manifest” or “to offer”: virtutem praestare ‘give proof of courage’, pietatem praestare ‘to prove one’s affection’; se praestare ‘to show oneself (as such)’. These uses evidently pave the way for the expression praestare pecuniam alicui ‘to put money at somebody’s disposal, to lend (French prêter) it to him’.

But we can understand that in this specialized sense praestare at the beginning, and for a long time, was applied to a loan without interest, a gracious offering, a testimony of good-will, and not a financial operation. Such a “loan,” which was simply an advance of money, is different from the loan called mutuatio, in which the notion of reciprocity appears, implying the exact restitution of what one has received, and is still further removed from fenus ‘loan at interest’.

The history of this notion considered in the different terms and in their separate development appears as an aggregate of complex processes, each of which achieved precision in the individual history of the separate societies. The problem everywhere is to establish what was the first value of these terms and how they became specialized in use. Even if some points of detail remain obscure, we have been able to show what the respective situation of the forms which came under consideration was, and under what conditions the extension or restriction of sense of certain terms came about.