Theft and robbery

(Inst.Gai.3.183.-208., lnst.4.1„ D.47.2. and C.6.2.)

We are concerned here with theft as a delict rather than as a crime. Thieves were often not worth suing, particularly if they had disposed of the property that they had stolen.

10.3.1 The essentials of theft (furtum)

Paul, Edict, book 39: Theft is a fraudulent interference with a thing with a view to gain, whether by the thing itself or by the use or possession of it. (D.47.2.1.3.) (cf. the definition in Inst.4.1.1.)

Theft consisted of several constituent elements, as emphasized in Paul's definition. Each needs consideration. It should be noted, however, that this definition of theft (and the related one in Inst.4.1.1.) are post-classical and it is a matter of contention whether all the elements were necessary requirements from the start. (See Watson, A., 'The Definition of Furtum and the Trichotomy' (1958) 28 TR, 197-210 on the evolution of the concept of theft in Roman law and Ibbetson, D., ‘The Danger of Definition: Contrectatio and Appropriation', in The Roman Law Tradition, 54-72, as well as Birks, Obligations, 159-60.)

W.3.1.1 'Interference'

There had to be—at least in classical law—a contrectatio by the thief, i.e. some physi­cal interference with (or handling of) the property. A thief normally takes away the property that he is stealing, and it seems that an act of removal was necessary in early law.

Pomponius, Sabinus, book 19: If, when my tame peacock escaped from my house, you chased it so that it disappeared, I could have the action for theft against you if someone else should take it. (D.47.2.37.)

There was no physical contact in this scenario, and yet it was theft because you were 'interfering' with my property.

What if a man took part of the whole, e.g. a shovelful from a heap of corn—had he interfered with the part only, or with the whole? The problem did not arise in early law: you stole only what you took (since removal was necessary). But once handling became the actus mis of theft it was arguable that a man stole whatever he interfered with, even if he removed only a part of it. Paul states:

Paul, Sabinus, book 40:...

It is not clear whether Paul is agreeing in this passage with Ofilius and Trebatius. The analogy with touching an ear is unconvincing: the ear is inseparable from the body whereas a heap of corn consists of separate grains. The most plausible expla­nation of this passage is that strictly there is theft of the whole heap, but that the thief is liable only for what he takes. The confusion is made worse by the applica­tion of different tests in other passages:

Paul, Submits, book 9: If a chest be broken into so that, say, pearls may be removed and they are handled with theftuous intent, it is only of them that theft may be held to be committed... The remaining things, set aside to get at the pearls, are not tampered with for the pur­pose of their theft. (D.47.2,22.1.)

This approach to the problem, linking the act of interference with the intent to steal, makes good sense; however, there is no clear indication in the texts which of the tests discussed was thought to be correct. Indeed, it seems that the test varied according to the circumstances and the nature of the thing stolen. Tor example, it would appear that a thing could not be stolen if it was too heavy to be moved:

Paul, Sabinas, book 40: If, again, a person open or break into something of too great weight to be removed, an action for theft will lie against him, not for the whole contents but only for what he removes, because he could not remove the whole thing.... (D.47.2.21.8.)

This passage suggests that asportation was still required in some cases of theft in the classical period, even if it was no longer necessary in all cases.

10.3.1.2 'Fraudulent'

The interference must have been fraudulent—there could be no liability unless the defendant acted dishonestly (cf. D.47.2.52.20.). So he had to be capable of dishon­est intent. It followed that the insane could not commit theft, except during a lucid interval; nor could children unless they were old enough to be capable of dishonest intent (cf. Inst.Gai.3.208.). The test of dishonesty was purely subjective—what mat­tered was the defendant's genuine belief. If he genuinely thought that he was enti­tled to take the property, he could not be liable, however inconceivable or unlikely his belief. This, at least, was the rule that prevailed in the classical period. The view of Sabinus that the defendant's belief must be founded on reasonable grounds was not generally accepted. Of course, the more unreasonable the belief, the more unlikely it would be that the judge would accept the defendant's story.

The interference had to be without the consent of the owner: if he consented, there could be no theft, however dishonestly the defendant had acted. But setting a trap for the thief, e.g. by leaving goods lying around invitingly, did not constitute consent. If the owner did not consent to the taking but the defendant genuinely thought otherwise, there was no theft. What of the converse case, i.e. where the owner consented but the defendant did not know this? This could not be theft: the defendant has been dishonest, but he has interfered with the property with the consent of the owner. Pomponius thought differently but his view did not prevail. In proving fraudulent interference it was not necessary to show, unlike the case with theft under the Theft Act 1968, that the defendant intended to deprive the owner permanently of the property.

What if the defendant showed repentance after committing theft? Apparently, it made no difference:

Ulpian, Gurule Aediles' Edict, book 1: One who appropriates another's thing with a view to his own gain is a thief, even if, changing his mind, he later returns it to the owner; no one ceases to be guilty by his own repentance over such a wrong. (D.47.2.66.)

This may be so, but it Is unlikely that a repentant thief would have been sued in such a case (i.e. where he restored the property), unless the owner had suffered some particular loss through being deprived of the property.

103.1.3 'A thing'

Things could not be. stolen unless they were capable of being privately owned. Hence, there could not be theft of things such as city gates or walls, for example. But woe betide you if you did 'interfere' with such things—a criminal sanction would almost certainly be imposed. To be stolen the thing had to be owned, or at least there had to be someone with rights in the property sufficient to sustain an action for theft. Thus, if the property had been abandoned, there could be no theft, however dishonest the defendant might have been:

Ulpian, Sabinin, book 41: If its owner has abandoned something, I will not commit theft of it, even though I take it with theftuous intent; for there can be no theft without an owner of the object... (D.47.2.43.5.)

What if the thing had not been abandoned, but the defendant had taken it believ­ing that it was abandoned? This could not be theft since the taking was not fraudu­lent, providing that the belief had been genuinely held.

Could land be stolen? After some controversy amongst the jurists (Sabinus, for example) the rule emerged in the classical period that land could not be stolen, but that things forming part of the land could be, e.g. fixtures, crops, minerals. Severing and removing such things could amount to theft, or handling them after their sev­erance by someone else; but not simply interfering with unsevered things:

Alfenus, Digest, Epitomized by Paul, book 4: If someone dig a hole to remove lime and he does remove it, he is a thief, not for digging but for taking.

Could human beings be stolen? Slaves could be, of course. The kidnapping of slaves was not uncommon: if the slave consented, he would be regarded as an accomplice. A slave who escaped from his master was regarded (with somewhat harsh logic) as having stolen himself, and as constituting stolen property. This had the important consequence of preventing the slave from being acquired by a third party through usucapio. Free persons too could be stolen, e.g. a child in potestas, or a wife in a manus marriage, the action for theft being available to the paterfamilias or husband respectively. (Consider the crime of kidnapping (plagium) in Roman criminal law, see Robinson, O. R, Criminal Law of Ancient Rome (1995), 32-5.)

10.3.1.4 'With a view to gain' (animus iucrandi)

The purpose of this requirement was to distinguish between theft and acts of wan­ton destruction (which could result in liability under the lex Aquilia). It was impor­tant to make the distinction because the delicts had different remedies:

Ulpian, Edict, book 37: If a man wave a red flag and cattle rush off to fall into the hands of thieves, assuming that he has a malicious intent, he is liable to the action for theft. But even if he did not act with theftuous design, he should not go unpunished for such pernicious con­duct; hence, Labeo writes that an actio in factum should be given against him. (D.47.2.50.4.)

There is little guidance in the texts on the meaning of 'gain', but it seems that it was widely construed and was not restricted to economic gain. For example, taking A's property to give it to B, thereby gaining favour with B, amounted to theft. However, merely obtaining a sense of satisfaction, or acting purely out of spite, would be ) insufficient to constitute gain.

A further problem with animus tucrandi is that it was not stated to be an essen­tial requirement of theft in the definitions contained in the Institutes of Gaius and Justinian. It is arguable that the requirement of an intention to gain was not part of classical law, but that it was added in the time of Justinian to differentiate between theft and wrongful damage under the lex Aquilia. According to this argument, the passages in the Digest concerning intention to gain must be interpolated. But was not the need to distinguish between the two delicts just as acute in the classi­cal period as in the late Empire? The requirement of animus lucrandi makes good sense—the most plausible view is that it was certainly part of late classical law. Why it was omitted from Justinian's Institutes remains unclear.

10.3.1.5 Theft of 'use' (furtum usus)

Theft could be committed not only by taking a thing dishonestly but also by unau­thorized use of a thing belonging to someone else. In the latter case, the normal scenario was that the wrongdoer acquired the use of the thing lawfully (through a contract) but then stole the use by misusing the property. So, if you lent a chariot to enable me to drive myself to the market to purchase groceries, and I used it instead to race in the arena, I committed theft if my conduct amounted to unauthorized use (however temporary) of the chariot. But unauthorized use could not amount to theft unless there was dishonest intent:

Pomponius, Quintus Mucins, book 38: One who uses a thing which he has borrowed or which was deposited with him otherwise than on the terms on which he accepts it, will not be liable fortheft, if he believe that he is not acting contrary to the owner's will.... (D.47.2.77pr.)

On theft of use, see Watson, A,, 'D.47.2.52.20: the Jackass, the Mares and "Furtum"' in Stud! Volterra II, 445-9.

10.3.1.6 Theft of 'possession' (furtum possessions)

If an owner of property dishonestly interfered with someone else's rights in that property, he committed theft—he had stolen the 'possession' of the property. The concept that a person could simultaneously be both the owner and the thief of the same property seems, prima fade, somewhat strange, but is one that can be readily justified in appropriate circumstances. For example, if A sold, a thing to B which A had pledged to C as security, A committed theft. Theft of possession could occur only if the victim had a right in rem in the property. (See Thomas, J. A. C., 'Furtum Pignoris: A Commentary on Commentaries', in Studi Sanfilippo I, 585-600 and Van den Bergh, G. C. J. J., 'Custodia and Furtum Pignoris', in Studi Sanfilippo I, 601-14, as well as Birks, Obligations, 180-7.)

10.3.2 Remedies

There were two types of civil remedy for theft: recovery of the stolen property, and the obtaining of damages. These remedies were cumulative, not mutually exclusive—a thief, if found liable, would have to return the property (or its equiva­lent) and pay damages, as a general rule (cf. Inst.Gai.4.4.),

10.3.2.1 Recovery of property

The standard proprietary remedies, i.e. vindicatio, the possessory interdicts, and in classical Roman law the actio ad exhibendum ('the action to display') were available against the thief or his heirs; or against anyone shown to be in possession of the stolen property, even if their possession had been acquired honestly, e.g. innocent third-party purchasers. The chief disadvantage of the vindicatio and the posses­sory interdicts was that they were not available if the stolen property could not be traced, or if it had lost its identity or had been destroyed. In such a case the condictio furtiva ('the action for stolen property') was allowed to the owner against the thief or his heir for recovery of the equivalent value. The condictio eventually became the most frequently used of these actions (assuming that the thief had been found) as it was not essential to prove the whereabouts of the property (compare D.13.1. and C.4.S. where this remedy is discussed in detail).

10.3.2.2 Damages

The action for damages was called the actio furti ('the action for theft'). It was a penal action in that the plaintiff could recover double or fourfold damages from the defendant, depending on the type of theft that had been committed, i.e. whether it was manifest theft or not. Additionally, the thief was subject to infamia. The poten­tial sanctions were certainly onerous in the civil law, but the thief had first to be caught—that might not happen very frequently in a big city without an adequate police force. And, even if caught, the thief might not be worth suing.

(a) Manifest and non-manifest theft The plaintiff could recover fourfold damages for manifest theft (furtum manifestum), twofold for non-manifest theft. Manifest theft was originally punished harshly under the Twelve Tables—a slave could be executed; a freeman could be enslaved by his victim. The praetors eventually replaced the original sanctions with the action for fourfold damages.

What was manifest theft? In early law it seems that the test was whether the thief had been caught in the act of stealing—if he had been, the theft would have been obvious, i.e. manifest. This may explain why manifest theft led to more severe sanctions than non-manifest theft: the former was considered more outrageous. If you catch someone burgling your home, your shock and sense of outrage will probably be much higher than if you discover the burglary after the bird has flown. Pugsley has controversially suggested that originally furtum manifestum consisted of the theft of res mancipi: the theft of such important property would be obvious in very early Rome (Property and Obligations, 28 ff.). In the classical period a wider test of manifest theft came to be applied:

Ulpian, Sabinas, book 41: But is a thief manifest only if he be caught in the act or also if he be apprehended elsewhere? The better view is that which appears in the writings of Julian, that is to say, that although he be not taken at the scene of the offense, he will still be a manifest thief if he be taken with the stolen thing, before he has taken it to its intended destination. (D.47.2.3.2.)

From this passage it seems that the timing of the arrest was crucial. If the thief was arrested after he had taken the property to his intended destination, the theft was no longer manifest, even if the thief was caught in possession. But an actual arrest was not always necessary—an attempted arrest might suffice in some circumstances:

Ulpian, Sabinas, book 41: Celsus, though, on the issue of apprehension, adds that if, whenyou saw the thief in the act and ran to arrest him, he made his escape by discarding his loot, he would be a manifest thief. (D.47.2.7.2.)

Moreover, it was manifest theft if stolen property was found in a man's premises following a ritual search. The ritual required that the searcher (presumably a male) should wear only a loincloth, and that he should carry a platter. The former require­ment lessened the possibility of planting property on the premises, while the plat­ter was possibly intended to carry an offering to the gods, at least in early times, or to provide a receptacle for the goods when found. This procedure, sanctioned by the Twelve Tables, became obsolete during the Republic. Gaius referred to it as an ancient practice, and as something of a joke—'lex tota ridicula est (Inst.Gai.3.193a.).

(b) Measure of damages What was it that was doubled or quadrupled under the actio furti? There is some inconsistency in the texts. Ulpian appears to take the view that it was the value of the property stolen:

Ulpian, Edict, book 37: In the action for theft, it is not the plaintiffs interest which is quadru­pled or doubled but the true value of the thing. (D.47.2.50pr.)

However, this text is contradicted by other jurists, and even by Ulpian himself in the following passage:

Ulpian, Edict, book 37: If a slave be stolen who has been named as heir in a will, the plaintiff, in the action for theft, will recover also the value of the inheritance, if the slave be dead before he could accept the estate at his master's direction.... (D.47.2.52.28.)

One can argue that there is no real conflict in these passages in that 'true value' surely encompasses 'the plaintiff's interest', but that is not what Ulpian appears to be saying in D.47.2.50pr, (see earlier). Moreover, the formula used in the actio furti suggests that what the plaintiff was seeking was double or quadruple the amount that the theft 'cost' him. Thus, the most probable view is that the measure of dam­ages was the plaintiff's 'interest' in the stolen property, and not the property's value. Certainly, consequential loss resulting from the theft could be recovered if the measure of damages was based on the plaintiff's interest, For example:

Celsus, Digest, book 12: If something be stolen from you which you had promised under pen­alty to deliver by a certain date with the result that you have to pay the penalty, that can be taken into the assessment in the action for theft. (D.47.2.68.1.)

The plaintiff was entitled to the highest value that his interest had been worth between the theft and the trial. Thus, if the interest had depreciated after the theft, its value at the time of theft was taken into account, not its depreciated value. If the property appreciated in value, then it was the higher value to which the plaintiff was entitled.

(c) The plaintiff Who could sue under the actio furti?

Ulpian, Sabinus, book 29: A person who has an interest in the thing not being stolen will have the action for theft. (D.47.2.10.) (cf. lnst.Gai.3.203.)

Behind this deceptively simple rule lie some difficulties. There were some persons who could not sue even though they could be reasonably said to have had 'an inter­est in the thing not being stolen'. For example, a creditor could not sue for theft of a thing from his debtor, unless it was pledged under the contract of pignus (see 9.5.4) (cf. Thomas, J. A. C., 'Furtum Pignoris' (1970) 38 TR, 135-62), even if there was no other effective security for the debt.

Who did have 'an interest in the thing not being stolen'? The jurists' approach was casuistic—they were content to list instances when someone could or could not bring the action, but they eschewed the formulation of general principles. However, it is possible to classify those entitled to sue as having either a positive or negative interest. A positive interest connotes that the plaintiff was owner, or in certain respects in the position of an owner, such as a bona fide possessor, or a usufructuary. The position of the usufructuary depended on which part of the usufruct was stolen. If the theft was of the fruits, then the usufructuary alone had the action (as long as he had gathered the fruits). But if the thief stole the actual property that was held on usufruct, the action was divided:

Ulpian, Sabimis, book 42: If a fructuary slave be stolen, both the fructuary and the owner have the action for theft. The action is thus divided between fructuary and owner; the fructuary sues for twofold the value of the fruits or for his interest in the slave's not being stolen; the owner for his interest in his property not being stolen. (D.47.2.46.1.)

A holder was said to have a negative interest, sufficient to enable him to bring the actio furti, if he had a contractual duty to compensate the owner for failure to return the property (such as the borrower, the conductor operis, and the mandatary). The owner was protected well: he had the proprietary remedies against the thief and was entitled to compensation from the holder for failure to return the property. However, if the holder sued the thief under the actio furti, the owner could not do so—the negative interest barred the positive interest. This appears to have given the holder of the negative interest an advantage if the theft was manifest: he could recover fourfold, whereas the owner recovered double, i.e. the thing (or its equiva­lent value) and the contractual compensation. But the holder had to find the thief.

What if the holder was insolvent? This made an important difference because in such a case the owner was unlikely to receive compensation from the holder. So the owner was given the actio furti:

Ulpian, Sabimis, book 29: And so a fuller who accepts garments for cleaning and attention will always have the action because he is liable for their safekeeping. But if he should be insolvent, the action reverts to the owner; for nothing is at the risk of one who has nothing to lose. (D.47.2.12pr.)

However, in the case of commodatum (see 9.5.2) Justinian altered the position. Whether the borrower was solvent or not, the lender was given the choice of suing the borrower under the contract of loan, or the thief under the actio furti. In the lat­ter case, the borrower no longer had a negative interest (since he could not be sued under the contract) and thus could not proceed against the thief.

The position of a buyer before delivery of the goods was unusual in that he was strictly neither a positive nor a negative interest holder. Until delivery the seller remained owner, the buyer having only a contractual right—he did not have a positive interest. Nor did he satisfy the criteria necessary to be a negative interest holder, i.e. contractual liability for failure to return property. Thus, if the goods were stolen before the seller had delivered them, the right to sue remained in the­ory with the seller; but it seems that the buyer could insist on that right being assigned to him. And if the seller himself sued the thief, the buyer was entitled to the fruits of the action, i.e. whatever the seller recovered. For other exceptional cases, see Buckland, Textbook, 580.

The interest of the plaintiff, whether positive or negative, must have been acquired without fraud: it had to be 'honest'.

(d) The defendant Who could be sued under the actio furtit The action lay against ; the thief but not against his heirs. Accomplices were fully liable; payment by one did not absolve the others. The basic rule was that an accomplice was someone who helped the thief by giving advice or physical assistance:

Ulpian, Edict, book 37: A person is deemed to be an accomplice, as adviser, who persuades, directs, and, by instruction, gets the theft committed; a man gives assistance, who provides aid or assistance at the actual taking of the goods. (D.47.2.S0.3.) It seems that originally both elements—physical assistance (ops) and advice (consilium)—had to be proved against the defendant accomplice, but in the classical period he became liable on either ground (see MacCormack, G., 'Ope Consillo Furtum Factum' (1983) 51 TR, 271-93 for a discussion on the origin of the distinction between theft and complicity, as well as the criteria used to establish the latter). Moreover, the accomplice must have been shown to have acted dishonestly.

The actio furti could not be brought against family members for domestic thefts as domestic authority sufficed. Consequently, husband and wife could not sue one another for theft committed during marriage; but the praetors introduced a special action, the actio rerum amotarum (‘the action for things taken away'), available after divorce for the recovery of compensation. It is possible that the action was con­fined to things that had been taken in contemplation of the divorce. Fathers and children-in-power could not sue each other for theft, except in respect of a son’s acquisitions on military service. Nor could the actio furti be brought against other members of the household, e.g. slaves, or freedmen. However, in all these cases a theft had been committed; so, although no action was possible against the princi­pal, an accomplice who was not part of the household could be sued.

10.3.2.3 Other remedies

A number of special actions, pre-classical in origin, were available in certain unu­sual circumstances (cf. Buckland, Textbook, 588 and Lee, Elements, 373-5). An act amounting to theft could constitute another delict, e.g. wrongful damage to prop­erty or insulting behaviour. In such cases, the plaintiff would have to choose the most favourable action—he could not recover twice over (or more) in different actions. In the Empire, criminal prosecutions for theft became increasingly fre­quent under the cognitio extraordinaria procedure (see generally 3.4):

Ulpian, Edict, book 38: It must be remembered that now criminal proceedings for theft are common and the complainant lays an allegation. It is not a kind of public prosecution in the normal sense, but it seemed proper that the temerity of those who do such wrongs should be punishable on extraordinary scrutiny. Still, if that be the party's wish, he can bring civil proceedings for theft. (D.47.2.93.)

It seems that judgment in such proceedings barred the actio furti, and vice versa.

10.3.3 Robbery (rapind)

(Inst.Gai.3.209, Inst.4.2, D.47.8)

Damages under the actio furti were not affected by proof that the theft had been committed violently. However, a praetorian edict in 77 BC introduced a special action for violent theft (robbery)—the actio vi bonorum raptorum (‘the action for goods violently taken'). It was a response to the mob violence of the disturbed con­ditions of the time and was aimed particularly at the activities of armed gangs. In due course this action came to be treated as a separate delict, although in reality it was little more than an adjunct of theft. See Birks, Obligations, 188-91.

10.3.3.1 Essentials

Robbery shared many of the substantive principles of theft:

Ulpian, Edict, book 56: And generally, it is to be said that wherever I could have the action for theft for something done by stealth, I will have the present action. (D.47.8,2.23.) There had to be a dishonest and forcible taking of property; if the taker acted in good faith, he was not liable. This rule appears to have exonerated those who violently enforced genuinely held claims. Not surprisingly, Marcus Aurelius decreed that bona fide claims should be forfeited if enforced in a violent manner. In the late Empire an unfounded bona fide claim, if pursued violently, would necessitate not only return of the property seized but also its equivalent value (cf. Buckland, Textbook, 584).

What degree of violence was required? The texts do not provide a clear answer. The most plausible view is that even a small amount of violence sufficed, whether committed by an armed mob or an unarmed man.

10.3.3.2 Remedies

Fourfold damages were recoverable, providing that the action was brought within a year, The value of the property that had been seized was included in the damages— the action was part compensatory (recovery of simple value), part penal (threefold penalty). If the plaintiff obtained fourfold damages, he was barred from bringing a proprietary remedy for recovery of the property seized. However, if he preferred, he could seek to recover the property and sue for threefold damages. If the action was brought after a year, only the simple value could be recovered. The measure of damages was based on the value of the property, not on the plaintiff's interest in it (unlike the case with theft). The action was available to the heir of the victim but not against the heir of the defendant.

Who could bring the actio vi bonorum raptorum?

Ulpian, Edict, book 56:... Hence, whether it be lent, let, or pledged to me, or deposited with me, so that I have an interest in its not being removed, or if I possess it in good faith or have a usufruct or other right in it, such that 1 have an interest in its not being forcibly taken, it must be said that I have the action under discussion... (D.47.8.2.22.)

The rule was thus identical to that in theft (see Inst.4.2.).

Robbery was tried by recuperatores, a body of lay jurors (see 3.2.2.3); condem­nation involved infamia. Criminal proceedings were possible as an alternative, it seems, to the civil action. As regards civil proceedings, the plaintiff had a choice between the action for robbery and the actio furti. If the robbery amounted to a manifest theft, the actio furti would be more attractive since it was not limited to a year and gave fourfold damages plus recovery of the property or its value. Only if the violent theft had not been manifest would the action for robbery be generally preferable. If the plaintiff brought that action, it barred him from suing under the actio furti. But the converse did not necessarily apply:

Paul, Edict, book 22: If the action for taking by force be brought first, the action for theft will be refused; but if the action fortheft be brought first, the other will lie to recover the balance available. (D.47.8.1.)

Thus, if the plaintiff brought the actio furti for a non-manifest violent theft, he could later bring the action for robbery to recover the balance to which he was entitled, y

io.4