Hostname: page-component-586b7cd67f-dsjbd Total loading time: 0 Render date: 2024-11-27T14:02:01.582Z Has data issue: false hasContentIssue false
  • English
  • Français

Liability for Animals in Roman Law: An Historical Sketch

Published online by Cambridge University Press:  16 January 2009

Bernard S. Jackson
Get access

Extract

The problems of reconstructing the development of liability for animals in Roman law are considerable. Each of the remedies is attended by major difficulties for the historian. We have the following:

(1) A Twelve Tables remedy for pauperies—an archaic term, the meaning of which is obscure;

(2) A Twelve Tables remedy for depasturation, of which there is no surviving quotation, but only a small number of allusions;

(3) An edictum on dogs and certain wild animals, undated and surviving in two versions with hardly any juristic interpretation;

(4) A “lex Pesolania” concerning dogs, the very existence of which has been questioned;

(5) The classical actio de pauperie, the basis, extent, and remedy for which have all been assailed by the interpolationists;

(6) The classical actio de pastu, evidently superseded by Justinian's day and thus leaving but few traces.

The Common Lawyer, to whom these remedies may reasonably evoke the types of damage represented by scienter and cattle-trespass and the special rules for dogs and ferae, will not be surprised to find, in addition, the application of general culpa principles by the extension of the actio legis Aquiliae.

I. The Decemviral Remedies

Some strands of ancient tradition maintain that the Decemvirs adopted an eclectic method, incorporating within the Twelve Tables both ancestral custom and foreign law.

Type
Articles
Information
The Cambridge Law Journal , Volume 37 , Issue 1 , April 1978 , pp. 122 - 143
Copyright
Copyright © Cambridge Law Journal and Contributors 1978

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Dionysius, 2.27.3; 10.57.5; see Jackson (1975) 23 A.J.C.L. 508–509.

2 Müller, (1965) X R.E.Suppl. 521, asserts that the name actio de pauperie does not appear in the Roman sources. He overlooks Inst.IV.9.1, praeter has autem aedilicias actiones et (scil., actio) de pauperie locum habebit, and Dig. 19.5.14.3, neque de pauperie… agi posse.

3 Haymann (1921) 42 Z.S.S. 357; Wenger, L., Die Quellen des römischen Rechts (Vienna, 1953), p. 367Google Scholar, n. 94; Müller (1965) X R.E.Suppl. 523; see also Modrzejewski, , Hommes et Bëtes (ed. Poliakov, Paris, 1975), pp. 83–84;Google Scholaridem (1976) 221 Accad.Naz.d.Lincei 191 (La Filosofia greca e il diritto romano).

4 As implied by Paul, Dig. 9.1.4., haec actio utilis competit et si non quadrupes sed aliud pauperiem fecit, and by the structure of Ulpian's introduction to the action in lib. 18 ad Ed. In Dig. 9.1.1. pr. (infra, note 6) he comments successively on those terms which, impliedly, go back to the ancient statute. A possible objection to this analysis has been removed by Lenel, Das Edictum Perpetuum (3rd ed., Leipzig 1927), p. 195, note 9, who holds Ait praetor (one might add pauperiem fecisse) in 1.3 to be interpolated, the action not being edictal. This defeats any argument to the effect that the terminology of the clause si quadrupes pauperiem fecisse dicetur is edictal rather than decemviral. Of course, this is not to deny that the form of the clause reflects the formula.

5 Plutarch, Solon 24:3. Xenophon, Hellenica II, 4, 41, attests only the clause concerning the biting dog. Hence it is not inconceivable that Plutarch's tetrapodos was modelled on quadrupes of the Roman action. Plutarch lived at Rome for many years, and was involved in the education of Hadrian. For the texts see E. Ruschenbusch, Solönos Nomoi (Wiesbaden 1965), 80 no. 35. Further, infra, note 52.

6 Dig. 9.1.1.pr.: Si quadrupes pauperiem fecisse dicetur, actio ex lege duodecim tabularum descendit: quae lex voluit out dari id quod nocuit, id est id animal quod noxiam commisit, aut aestimationem noxiae offerre. The cause id est id animal quod noxiam commisit is an explanatory gloss; cf. Haymann (1921) 42 Z.S.S. 372; Biondi, , Actiones noxales (Palermo 1925), p. 15Google Scholar; Lenel (1927) 47 Z.S.S. 7.

7 As Vinogradoff, , Outlines of Historical Jurisprudence (Oxford 1922), ii. 194–195Google Scholar, was willing to do.

8 For the underlying methodology and assumptions, see Jackson (1974) 18 J.J.P. 77 et seq.; Essays in Jewish and Comparative Legal History (Leiden 1975), pp. 135Google Scholaret seq.

9 As did the author of P.S. 1.15.1, following the emendation of Cujas; infra, n. 52.

10 Despite the term's rarity and archaic flavour, Kerr–Wylie, Studi Riccobono, iv. 465, denied that it belonged to the XII Tables, and Müller, (1965) X R.E. Suppl. 521, regards the question as unsettled. Their argument depends in large measure upon animal quod noxiam commisit in Dig. 9.1.1.pr. and noxia autem est ipsum delictum in 1.2. But the former is interpolated (supra, note 6), and the latter is merely explanatory of aestimationem noxiae in what immediately precedes.

11 (1970) 17 R.I.D.A. 361–362. To the earlier etymological arguments add Robbe (1932) 7 R.I.S.G. 330 note 3.

12 P.S. 1.15.1. Fliniaux, Mëlanges Cornil, pp. 268–272, argues from the I.P. and L.R.B. that quidve depasta sit was absent from the original text.

13 The whole of P.S. 1.15 is a hasty collection of diverse remedies concerning animals (infra, note 10). Despite the gloss to Dig 19.5.14.3, Sed quid si mea pecora in tuo glandes tuas pasta fuerunt sine mea immissione? Respondit habet locum forte de pauperie Accursius, Fliniaux, Mëlanges Cornil, i. 281 note 4, is correct in denying, for classical times, the availability of the pauperian action where depasturation occurred sine immissione (cf., for the result, Kerr–Wylie, Studi Riccobono, iv, 465) because of failure to fulfil the contra naturam requirement, though he leaves the question open for the period of the XII Tables. For deliberate depasturation, the pauperian action would be barred also on the grounds of culpa; cf. Dig. 19.5.14.3. (infra, note 34, though the reason there is not expressed; it could be simply that the action still did not apply to damage to land or crops); Fliniaux, i. 256.

14 Bodily injuries are contemplated in Dig. 9.1.1.4–5; h.t.1.7; 2.1; 3. Note that in the two cases where damage to movables is likely, since an animal upsets the wagon it is pulling (h.t.1.4. and 1.9) Ulpian appears to consider only the ensuing bodily injury. Similarly, Alfenus restricts his attention to the killing of the slaveboy in Dig. 9.2.52.2. The one jurist who provides clear evidence of the availability of the action for damage to movables is Paul, Dig. 9.1.2.pr. For injuries to other animals, see Dig. 9.1.1.11 (Ulpian reporting Q. Mucius); cf. Dig. 9.2.57 (Javolenus), where the action is, in the circumstances, denied.

15 Laws of Eshnunna, §§ 53–57; Laws of Hammurabi, §§ 250–252; Exodus 21:28–32, 35–36; on which see, most recently, my Essays in Jewish and Comparative Legal History, pp. 108–152.

16 Dig. 9.2.52.2 (puerum cuiusdam); 9.2.52.3 (emptoris servus); 9.1.5 (agasso).

17 Dig. 9.1.1.11.

18 Dig. 9.1.1.4. The aliquem in aut si plus iusto onerata quadrupes in aliquem onus everterit haec actio cessabit damnique injuriae agetur would in isolation certainly suggest the inclusion of free men; and the text would be evidence for Republican times not only of the conceivability of a pauperian remedy for injury to a free man, but also of the availability of an Aquilian remedy. But this possibility appears to be negated by Dig. 9.2.11.5, where a similarly ambiguous aliquem must be restricted to a slave unless Julian is taken, in the next clause, to impose Aquilian liability for a deliberately inflicted injury to a free man (cf., most recently, Macqueron, Flores Legum Scheltema, pp. 147–148, who does not, however, object to alicui in Dig. 9.1.1.5; MacCormack (1975) 41 S.D.H.I. 23). It may be noted that both these ambiguous uses of aliquis occur in Ulpian lib. 18 ad Ed., in reporting the opinion of an earlier jurist; and the same is true of Dig. 9.2.7.7, where Ulpian appears to attribute Aquilian liability for injury to a free man to Celsus (but where the victim in his analogy is a puer). The loose use of aliquis in such contexts may thus be a stylistic trait of Ulpian rather than the result of compilatorial interpolation (as maintained by von Lübtow, , Untersuchungen zur lex Aquilia de damno injuria dato (Berlin 1971), pp. 119–120.Google Scholar This view follows in part from holding Dig. 9.2.13.pr., Liber homo suo nomine utilem Aquiliae habet actionem, etc., to be interpolated. There is no internal reason to do so.) But the argument for early Aquilian liability for injury to free men does not fall entirely with the Ulpianic texts. Independent, but equally ambiguous evidence is provided by Dig. 9.2.52.1, Alfenus lib. 2 Dig. (quidam praeteriens), on which see Lawson, Negligence in the Civil Law, p. 131, allowing that the text may indicate such liability as early as the late Republic.

19 Cf. Kerr-Wylie, Studi Riccobono, iv.465.

20 Dig. 9.1.3.

21 Dig. 9.2.2.pr.

22 Sauvage, A., Étude de thèmes animaliers dans la poésie latine (Latomus 1975), p. 12Google Scholar, observes that the term quadrupes is (in poetry) especially associated with the horse, and is un mot noble.

23 Cf. Ulpian, Dig. 9.2.29.6: Hac actione ex hoc legis capite (viz. ch. 3) de omnibus animalibus laesis, quae pecudes non sunt, agendum est, ut puta de cane.

24 The usual translation “panther” (e.g., Scott, Lawson) may not be quite accurate. It appears that the animal so described is either a leopard or a cheetah. By the Empire, pardus or leopardus were more often used for leopard, and thus Gaius' panthera (unlike that of the aedilician edict) may well be a cheetah. In the Loeb edition, C. H. Rackham translates Pliny's pardos, pantheras (VIII.xvii.41) “leopards, panthers” but notes that the panthera is really a large Indian leopard. See also Wotke (1949) XVIII, 3 R.E. 747–750; Toynbee, J. M. C., Animals in Roman Life and Art (London 1973), p. 82.Google Scholar The best account of the problem remains Jennison, G., Animals for Show and Pleasure in Ancient Rome (Manchester 1937), 183–187Google Scholar, who observes that “the panther does not exist in English zoological nomenclature.”

25 Dig. 9.2.2.2. The reason may possibly be interpolated. At any rate, the decision is consistent with the exclusion of bestiae, given Ulpian's understanding of that term: bestias autem accipere debemus ex feritate magis quam ex animalis genere: nam quid si leo sit, sed mansuetus, vel alia dentata mansueta? Dig. 3.1.1.6, the context being one who hires himself out ut cum bestiis depugnaret.

26 Gai., Inst. 2. 14a–16.

27 Yaron (1968) 36 T.v.R. 60–64; Daube (1970) 20 Buffalo Law Review 42; Jackson, Essays, pp. 162–166.

28 Some scholars suggest an original restriction to res mancipi. See Müller (1965) X R.E. Suppl. 523–524. Cuq, Manuel, p. 596, restricts the action to animals which assist man pour la culture. See also Lebigre, A., La Responsabilité pénale en droit romain classique (Paris 1967), p. 19.Google Scholar

29 Dig. 9.1.1.2. Cf. P.S. 1.15.3: feram bestiam vel quamcumque aliam quadrupedem.

30 Cf. Augustine, quoted infra note 32, illustrating the duality of the usages of both quadrupes and bestia; see also Robbe's comment (1932) 7 R.I.S.G. 351, note 1.

31 Thus Ulpian contemplates the availability of the action where the animal is a dog (Dig. 9.1.1.5), a bear (Dig. 9.1.1.10), and a ram (Dig. 9.1.1.11). On the inclusion of ferae in classical law, see further infra, text at nn. 91–98.

32 De Genesi ad litteram imperfectus liber, 15, at C.S.E.L. XVIII, III, 1, p. 496: Cum autem in latina lingua nomine bestiarum omne irrationabile animal generaliter significetur, hic tamen distinguendae sunt species, ut quadrupedes accipiamus omnia iumenta, serpentes omnia repentia, bestias vel feras omnia quadrupedia indomita, pecora vero quadrupedia, quae non operando adiuvant, sed dant aliquem fructus pascentibus, quoted by Robbe (1932) 7 R.I.S.G. 350–351. Though we may agree with Robbe (at pp. 348–359) that quadrupes normally bore restrictive connotations, and that such restrictions were originally reflected in the law, we cannot follow him in his view that this remained true throughout the classical period, so that animal is interpolated throughout Dig. 9.1. Equally unconvincing are the arguments of Ashton-Cross (1953) 11 C.L.J. 397–399, that there was no pauperian liability for wild animals in classical law. Against Ashton–Cross, see Nicholas [1958] Acta Juridica 187–189.

33 Cf. Macqueron, Flores Legum Scheltema, p. 133, note 1.

34 Dig. 19.5.14. 3: Si glans ex arbore tua in meum fundum cadat eamque ego immisso pecore depascam: Aristo scribit non sibi occurrere legitimam actionem, qua experiri possis: nam neque ex lege duodecim labularum de pastu pecoris (quia non in tuo pascitur) neque de pauperie neque de damni injuriae agi posse: in factum itaque erit agendum. On the text–critical problems, see Fliniaux, Mëlanges Cornil, i.254, notes 1–2.

35 Indeed, little has been added to our knowledge of the action at any period of its history since Fliniaux gave the matter his detailed, and on the whole sound, consideration in 1926. For his reconstruction of the action at the time of the XII Tables, see Mëlanges Cornil, i.280–288. His reasons for reading later evidence back to this period are not always clear.

36 Fliniaux, Mëlanges Cornil, i.281, basing himself on immisso pecore in Dig. 19.15.14.3, Dig. 10.4.9.1 (though the context concerns the actio ad exhibendum), and noting Dig. 50.16.30.3. Cf. Kerr-Wylie, Studi Riccobono, iv.475.

37 Finiaux, i.284–286, principally on the grounds that liability was based on the owner's own act. For the actio de pastu there is no extant statement comparable to that of Ulpian, Dig. 9.1.1.pr., concerning the actio de pauperie, except P.S.1.15.1 which, whatever its significance for later law (infra, p. 137), purports to tell us nothing about the XII Tables. For the earlier literature (from Cujas on the one side and Godefroy on the other), see Fliniaux, i.251–253. Lenel in the 3rd edition of Das Edictum Perpetuum, p. 198, abandoned his earlier view that the action was noxal, but Kaser, Das römische Privatrecht, 2nd ed. (1971), i.162, note 67, still inclines towards an original noxality.

38 XII T. VIII.9, per Pliny, N.H. XVIII. iii.12. See Fliniaux, i.272–274.

39 Fliniaux, i.283–284, maintains that the actio de pastu was excluded even when depasturation of frugem aratro quaesitam occurred by day, such circumstances falling within the actio furti. He thus arrives at a restriction of the actio de pastu to depasturation of natural, as opposed to cultivated, fruit and vegetation. But the texts cited for the actio furti (Dig. 47.2.21.pr.; 9.2.27.25) do not refer to animals. As Fliniaux admits (p. 283, note 3) the delictual act is secare, and his only reason for including depasturation is that in the special regulation for nocturnal depasturation of cultivated fruits the XII Tables put secare “sur le mëme pied que le pascere.”

40 Ex. 21:35; see Jackson (1974) 18 J.J.P. 74–77; Essays, pp. 130–133.

41 Ex. 22:4 (MT); see Jackson (1974) 25 J.Jew.St. 129–130; (1976) 27 138–141.

42 The arrangement of the fragments in the XII Tables is largely conjectural, nor may we rely on the arrangement of the Edict, as reconstructed by Lenel from Book 18 of Ulpian's commentary thereon (see also Fliniaux, i. 364–365) for a decemviral collocation of the actio de pauperie and the actio de pastu. But such an articulation of structure through the arrangement of the material is not vital. We find it in neither Hammurabi nor Plato nor (considering the small scale of the collection) the Biblical “Covenant Code” (Ex. 21–22).

43 P.S. 1.15.1; Rotondi, G., Leges Publicae Populi Romani (Milan 1912), p. 472.Google Scholar Even if the lex is genuine, its title is unlikely to have included de cane, as is often implied. In the context the words are clearly descroptive of the statute's effect. Thus Rotondi has Lex Pesolania(?) (de pauperie).

44 Dig. 21.1.40–42; Inst. IV.9.1. The conventional title occurs in neither of these sources and is justified only on the basis of P.S. 1.15.3, Feram bestiam… praetor prohibet. Though this latter source may be accepted as referring to the aedilician edict it is on all accounts a loose paraphrase upon which little reliance is justified. Nicholas, [1958] Acta Juridica 186 note 12, describes it a “garbled reference.”

45 Flores Legum H.J. Scheltema Oblati (Groningen 1971)Google Scholar, in which he discusses the lex Pesolania at pp. 136–137, and (1971) 18 R.I.D.A. 782–783, summarising a paper delivered at the S.I.D.A. Congress of Vienna 1970.

46 Flores Legum Scheltema, p. 137 note 13. In (1971) 18 R.I.D.A. 783, he seems prepared to date it a little later.

47 On the grounds that dogs were not protected by chap. 1.

48 I am informed by Dr. Graeme Barker of the University of Sheffield, who has studied animal domestication in Italy ((1976) 3 Journal of Archaeological Science 71–81, though not here dealing with the dog), that there is evidence of the domestic dog in Italy from neolithic times onwards. As for historic times, see the 4th cent. B.C. coin from Panormus, in Hilzheimer (1932) 6 Antiquity P1.1, facing p. 416. I am indebted to Dr. David Ridgway, of the University of Edinburgh, for this reference.

49 According to the criteria adopted above, rams were also beyond the original ambit of quadrupes. The underlying principle may have been that the farmer was responsible for those animals which were of direct benefit to him.

50 No indication of special legislation is preserved in the case of the ram, which Q. Mucius includes within the action in Dig. 9.1.1.11, and, according to Varro, Atticus certainly regarded the dog as a quadruped: Rerum Rusticarum, II.ix.1, Relinquitur, inquit Atticus, de quadrupedibus quod ad canes attinet. The Republican jurists were certainly capable of coping with such extensions, more so perhaps than their successors. See , Watson, Limits of Juristic Decision in the Later Roman Republic (Edinburgh 1969)Google Scholar; (1969) 37 T.v.R. 351–368; Law Making in the Later Roman Republic (Oxford 1974), pp. 116, 119Google Scholar, where he suggests that the XII Tables were accorded especially free interpretation at the hands of the late Republican jurists. But this does not of itself reveal how legal development proceeded in the third century B.C. and earlier.

51 Paul in Dig. 9.1.2.1 considers whether, in the circumstances of the case under consideration (on which see Macqueron, Flores Legum Scheltema, pp. 137–140), there may be an action canis nomine, which means, in the light of his actio utilis for non-quadrupeds in Dig. 9.1.4, that he regarded the dog as a quadruped. Similarly, Ulpiart in Dig. 9.1.1.5 (on which see Macqueron, pp. 140–147) considers without hesitancy whether haec actio (viz., de pauperie) is available to the victim of a dog which has given its master the slip.

52 The author of P.S. 1.15.1 is not suggesting either that the XII Tables (to which, in fact, he does not refer) regulated dog-bites, or that Solon's law covers all quadrupeds (though Plutarch does use tetrapodos, supra, note 5). He merely compares the Roman sanction with that of Solon. Indeed, if the quod of quod etiam lege Pesolania de cane cavetur is restricted to its immediate context (aut quadrupedem dedat) the comparison should satisfy even those who would deny the availability of the alternative (payment of compensation) in Solon's law. Thus the author is merely saying “which sanction is also provided by a law of Solon concerning the dog.” Macqueron, Flores Legum Scheltema, p. 136, note 12, objects on the grounds that the law attributed to Solon (on the basis, according to Cujas, of Plutarch 24.3 and Plato, Laws, 936a, though the latter does not mention Solon) does not deal de façon particuliëre with damage caused by dogs. But Plutarch stresses the rule concerning dogs as a special feature of Solon's law. The existence of such a dog-law is supported by Xenophon, Hellenica II, 4, 41, though he does not mention its source.

Macqueron's text-critical argument, at pp. 136–137, is also questionable. He relies upon the Interpretatio and its agreement with an earlier version (referring, presumably, to Lex Rom.Burg. XIII, 1). For a synopsis of the texts, see Fliniaux, Melanges Cornil, i,269. The Interpretation reads quod etiam de cane similiter est statutum, and the Lex Rom.Burg. XIII, 1, quod etiam de cane et bipede placuit observari, secundum speciem Pauli Sententiarum. The latter text as well as the former interprets but does not quote a text of the Sententiae, and there is no reason why the text thus interpreted should not have been that preserved for us in the Breviary. Thus, an original reference to a lex Solonia (going back, perhaps, to an original work of Paul) was misunderstood or amended by the more practically-oriented compiler of the Sententiae, and the resultant lex Pesolania was interpreted as a genuine statute by the author of the Interpretatio. According to Macqueron, the unknown name of an otherwise unattested statute was altered by the copyists to the name of a magistrate, Pesolanius, who (as he concedes) in equally unknown. For variant readings Pesulania and Pesolonia, see Cujas, Opera (Venice 1758), i.332.

53 The availability of the actio legis Aquiliae goes back to Proculus, according to Dig. 9.2.11.5. Macqueron, Flores Legum Scheltema, pp. 148–149, doubts that the actio directa can have been granted by Proculus. But see MacCormack (1975) 41 S.D.H.I. 14–17.

54 Dig. 9.1.2.1.

55 Rerum Rusticarum II.ix.7, Sequitur quartum de emptione… De sanitate et noxa stipulationes fiunt eaedem, quae in pecore, nisi quod hic utiliter exceptum est. At II.vii.6 it is said that the sale of horses proceeds in a fashion similar to that of oxen and asses, which change ownership according to the Manilian actiones, and in II.v.11 the stipulation (there too implied to be Manilian) is given as illosce boves sanos esse noxisque praestari. Of the sale of asses Varro says de sanitate ac noxa solet caveri, II.vi.3; for pigs, see infra, note 66.

56 , Storr-Best, Varro on Farming (London 1912), p. 19Google Scholar, translates II.v.11 noxisque praestari “and that the buyer incurs no liability for damage done by them?” But in a note to a parallel passage, II.iv.5, at p. 166, he inclines, on poor evidence and without noticing the texts on pauperies, to “warranted free from dangerous habits.” Tilly, B., Varro the Farmer (London 1973), p. 256n.Google Scholar, sees in the phrase a reference to noxal surrender.

57 (1972) 7 I.J. 138–150, esp. 144, for the stylistic association.

58 , Lenel, Das Edictum Perpetuum (Leipzig 1927), p. 566Google Scholar, though taking the view that the clauses from si adversus ea in Dig. 21.1.42 are a close paraphrase of the original. Daube goes further, in arguing from other examples of the form that no specific sanctions were originally prescribed. See Forms of Roman Legislation (Oxford 1956), Chap. 3 esp. p. 44Google Scholar; (1961) 78 Z.S.S. 390–391; The Classical Tradition, Literary and Historical Studies in Honor of Harry Caplan (Ithaca, 1966), pp. 222–231.Google ScholarCf., more recently, , Watson, Law Making in the later Roman Republic (Oxford 1974), 85–86.Google Scholar

59 For such traders, see Jennison, G., Animals for Show and Pleasure in Ancient Rome (Manchester 1937), p. 45Google Scholar, in commenting on Plautus, Poen. 1011–1012. On the problem, see further infra at n. 74.

60 Pliny, N.H. VIII, 17, 64, and the discussion of Jennison at pp. 47–48.

61 There is no reliable evidence of pauperian liability before Gaius; supra at nn. 16–20.

62 Cf. Watson, Law Making in the Later Roman Republic, p. 86.

63 That the actio de pauperie and the aedilician actions were concurrent is attested certainly for later times by Inst. IV.9.1.

64 The Institutes of Justinian (Amsterdam & Oxford 1975), p. 307Google Scholar, rendering verrem, aprum, “boar, domesticated or tame.” For “tame” we are presumably intended to read “wild.”

65 Dig. 9.2.2.2: sed an sues pecudum appellatione continentur, quaeritur: et recte Labeoni placet contineri.

66 Rerum Rusticarum II,iv,5: emi solent sic: “illasce sues sanas esse habereque recte licere noxisque praestari neque de pecore morboso esse spondesne?” That verres is included here as the male of the species is clear from II,iv,4: boni seminis sues animadvertuntur a facie et progenie et regione caeli: a facie, si formosi sunt verris et scrofa… In this case the stipulation is not attributed to the Manili actiones.

67 Thomas, The Institutes of Justinian, p. 308. It should, however, be noted that while the Digest purports to contain a direct quotation, the Institutes do not. Moreover, the most natural place to find additions to the list is at the end, and this is not where the wolf or the panther occur. The conclusion thus appears to be that the list did attract additions, but these are not necessarily revealed correctly by comparison of the Digest and Institutes.

68 Varro identifies two genera of dogs, the one used for hunting, the other a guard dog. The latter was used to guard both sheep etc. and the homestead. See Rerum Rusticarum II.ix.1–2, I.xix.3. It may be noted that the sheep–dog in antiquity was trained to follow (R.R. II.ix.5) and protect the sheep, but not to herd them. On hunting dogs, see Aymard, J., Essai sur les chasses romaines (Paris 1951)Google Scholar, ch.XII; on dogs generally, Keller, O., Die antike Tierwelt (Leipzig 1909), i.91–151Google Scholar; Orth (1913) VIII R.E. 2540–2582; Toynbee, Animals in Roman Life and Art, 102 et seq.

The ability of the verres to defend itself against a wolf is noted and explained by Varro, R.R. II.ix.I, in suillo pecore tamen sunt quae se vindicent, sues verres, maiales, scrofae, prope enim haec apris, qui in silvis saepe dentibus canes occiderunt.

69 Boars were sometimes kept in leporaria for hunting purposes: Varro, R.R. III.iii.3; Toynbee, pp. 131–132. Jennison, p. 43, suggests that they appeared in venationes from a very early period, citing Varro, R.R. III,13. For the classification of aper as a fera see Ulpian, Dig. 9.2.29.6.

70 Macqueron (1971) 18 R.I.D.A. 783 appears to hold that the edict at first included only dogs. But if both the farm animals, canis and verres, were there from the beginning, it is easy to understand how aper might next Have been attracted. On the farm use of the verres see note 66 supra; Varro, R.R. II.iv.3, 4, 7, 21, 22.

71 Modrzejewski (1976) 221 A.N.L. 189. I have not encountered evidence of private collectors before the 1st century a.d. Epictetus (c. a.d. 50–120), IV.1.25, records that “men shut up tame lions in a cage, and bring them up, and feed them and some take them around with them.” In the third century, the Emperor Elagabulus appears to have made a habit of letting his wild beasts loose on his guests, at table or in their beds: Script.Hist.Aug., Elagabulus 21,1; 25.1, and Aurelianus in a.d. 273 presented to private individuals the elephants and other ferae mansuetae which had graced his triumph “that the privy purse might not be burdened with their food,” S.H.A. Aurelianus 33,4. See also Jennison, p. 132.

72 In 186 b.c. (after his tenure of the consulship, if Livy's date is correct) M. Fulvius Nobilior put on a venatio with lions and pantherae, Livy, 39.22.2; in 169 b.c. the aediles P. Cornelius Scipio Nasica and P. Lentulus displayed 63 Africanas (prob. leopards), 40 bears and an unspecified number of elephants, Livy 44.18.8. Later aedilician beast shows include those of Q. Mucius Scaevola (no less) and Lucius Licinius in 104 b.c., Pliny N.H. VIII.20.53; Domitius Ahenobarbus, who displayed 100 Numibian bears in 61 b.c., Pliny VIII.24. On the attempts of M. Caelius Rufus, who was standing for the aedileship in 50 b.c., to badger his friend, Cicero, then governor of Cilicia, into sending him pantherae for his victory games, see Toynbee, Animals in Roman Life and Art, p. 20. See generally Jennison, pp. 42–59.

73 Law Making in the Later Roman Republic, p. 86.

74 Supra, at note 59.

75 Toynbee, p. 93, notes that it was imported, e.g., from Scotland: Martial, De Spectaculis 7.3 (1st century a.d.), Caledonicus ursus. For imported bears in the ludi, see Livy, 44.18.8; Pliny, N.H. VIII.55.132, supra, note 72. See also Seneca, De Ira II.31.5–6. Cf. Jennison, p. 49. But the imported bear seems to belong to a period later than that of the edict.

76 Toynbee, p. 93.

77 On bear hunting, see Aymard, J., Essai sur les chasses romaines (Paris 1951), p. 13Google Scholar esp. sources cited note 7 (Bibl. des Écoles Françaises d' Athènes et de Rome, 171). In both cases, the hunting could have been in the context of the venationes, in their early form. See Jennison, p. 43.

78 Ashton-Cross (1953) 11 C.L.J. 397. But we do not know how the words ut cuiquam nocere damnumve dare possit were interpreted.

79 Cf. Nicholas [1958] Acta Juridica 185; Lebigre, La Responsabilité pénale, p. 20 note 2.

80 The existence of a remedy for fatal injuries is stressed in P.S. 1.15.2. The reference to solidi is interpolated: Honoré (1972) 7 I.J. 144; Watson, Law Making in the later Roman Republic, p. 85 note 6.

81 Supra, at notes 18–20.

82 Assuming this to be comprehended within the concluding ceterarum rerum, quanti damnum datum factumve sit, dupli, Dig. 21.1.42. On aliudve quod noceret animal in Dig. 21.1.41, see infra, note 98.

83 Supra, note 14.

84 Dig. 9.1.1.11.

85 Or at least were not kept gregatim. Rams are not included in the list at Dig. 9.2.2.2; supra, p. 125, and oves may not include them. The list continues with caprae, she-goats.

86 Supra at n. 55.

87 Supra, note 18.

88 Dig. 9.2.52.2 (Alfenus); Dig. 9.1.1.4 (Ulpian reporting Servius; Ulpian himself would have given an actio in factum, Dig. 9.1.1.7); Dig. 9.2.9.13 (Ofilius). See also Dig. 9.2.57 (Labeo); Dig. 9.2.11.5 (Proculus); Dig. 9.2.8.1. (Gaius). On the distinction between the actio directa and the actio in factum in some of these texts, see Watson (1969) 37 T.v.R. 353–355.

89 If a slave or filius familias was at fault in respect of the animal, he could, presumably, be noxally surrendered.

90 And culpa, we are told, included imperitia or infirmitas, Dig. 9.1.8.1 (Gaius). See Lawson, Negligence in the Civil Law, arguing for the genuineness of the text, against Haymann, Rotondi, and Kunkel.

91 Principally, those of Dig. 9.1.1.10 and Inst.IV.9.1. See, most recently, Modrzejewski (1976) 221 A.N.L. 181–183.

92 Taking his expansive interpretation of quadrupes in Dig. 9.1.4 against the background of the interpretation of quadrupedemve pecudem as excluding bestiae… veluti ursi leones pantherae, in Dig. 9.2.2.2.

93 Dig. 9.1.1.10.

94 Inst. IV.9.pr. (despite the reiteration of Ulpian's quondam dominus argument); Dig. 9.1.1.10, In bestiis autem propter naturalem feritatem haec actio locum habet, which may be regarded as interpolated. Note the difference in the usage of bestia from Ulpian's own in Dig. 3.1.1.6 (supra, note 25). As for the concurrence of the actio de pauperie and the aedilicias actiones in Inst. IV.9.1, we need only limit the latter to the canis and verres.

95 Inst. IV.9.pr.

96 Dig.9.1.1.7; Inst. IV.9.pr. More of this on another occasion.

97 Supra, note 71.

98 Dig. 21.1.41. Indeed, the extension goes beyond animals classified as ferae, as Ashton-Cross [1959] C.L.J. 191, observes. The skill of the cento lies in the fact that the passage from Paul, in its original context, could well have meant “or generally any other harm caused by the animal…,” the aliudve quod noceret being an interpretation of ceterarum rerum (or its equivalent) in the text. On the technique, see , Daube, Festschrift Hans Lewald (Basel 1953), pp. 27–34Google Scholar; (1959) 76 Z.S.S. 259–261; Flares Legum H.J. Scheltema Oblati (1971), pp. 45–46.Google Scholar

99 See esp. Fliniaux, Mëlanges Cornil, i.288–294.

1 Dig. 10.4.9.1; 19.5.14.3; cf. 50.16.30.3.

2 Cf. the restriction of the English cattle-trespass to averia. In the context of novel disseisin, Bracton echoes the terminology of the actio de pastu: vel pecora immittendo, f. 161b.

3 Dig. 19.5.14.3.

4 The lack of direct evidence is attributable to the fact that an action ex sententia legis Aquiliae was allowed from a.d. 294 where depasturation occurred per iniuriam; C.3.35.6, and see Fliniaux, pp. 265–266, 293–294; Kerr-Wylie (1934) 51 S.A.L.J. 182.

Ulpian, Dig. 19.5.14.3, had already held that an actio in factum was available where the defendant had his animals consume fruit which had fallen on to his own land. According to the XII Tables, VII.10 (Pliny, N.H. XVI.15) the owner of a glans which fell on to a neighbour's land was entitled to collect it. On the interpretation of glans in this context, see Watson, Limits of Juristic Decision in the Later Roman Republic, p. 12.

5 E.g., Voet ad Pand. 9.1.1. To those listed by Fliniaux, pp. 251–252, add more recently Ashton-Cross (1953) 11 C.L.J. 401, 402; Modrzejewski (1976) 221 A.N.L. 179. Contra, Müller (1965) X.R.E. Suppl. 523.

6 Mélanges Cornil, i.268–270.

7 At pp. 271–272. Thus, it is argued, they avoided suggesting that a dog might “depasture” land. But this argument assumes too readily that the quod of the final clause refers to the whole of the preceding fragment. See supra, note 52.

8 P. Oxy. 34, 2704, at 87–88, where the prefect Titius Honoratus threatens herdsmen whose beasts spoil the crop with confiscation of the animals and “strong measures.” In the light of the edict's stated purpose, and despite the use of the verb epaphienai (which is weaker than the Roman immittere), it is likely that the edict was directed against depasturation however occasioned. It may, however, be that the “strong measures” would be invoked only where the herdsmen were at fault, as is suggested by comparison of two early-fourth-century petitions of Isidorus of Karanis: P.Mert. II.92 and P.Cair.Isid. 78, for which see now Modrzejewski (1976) 221 A.N.L. 194–197. See also the decretum of an African municipal council of a.d. 186, partially preserved in a Tunisian inscription, which cites an imperial rescript (apparently of local effect) concerning slaves who commit acts of depasturation with their masters' animals: Fliniaux, pp. 275–280. It may be noted that the remedies in the petitions are not strictly confiscation but division of the proceeds of sale of the animals between the (municipal) Treasury and the landowner (P.Mert. II.92) and sale at public auction for the benefit of the municipal treasury, the owner being protected in an undisclosed fashion from loss (P.Cair.Isid. 78). The Merton papyrus itself recites the fact that the headmen “frequently” issued proclamations on the subject; the possibility of variation as to the particular form of the sanction which deprived the owner of his animals is therefore not excluded.

9 Cf. Lenel, E.P., 3rd ed., p. 198, retracting his earlier view.

10 Thus 1.15.1 in its present form combines references to the actio de pauperie, the actio de pastu, and the “lex Pesolania.” 1.15.1a (this and 1.15.1b in the F.I.R.A. edition being found only in the Lex Romana Burgundionum) provides a remedy for damnum caused by a fierce dog not kept on its leash in the public highway—the result, apparently of detaching the dog from the edictum de feris (summarised in 1.15.2), and implying a form of Aquilian remedy; 1.15.1b concerns one who allows his diseased pack-horse or other animal to mix with others and infect them; 1.15.2 gives a version of the edictum, substituting the praetor for the curule aedile and feram bestiam for the original list, and omitting to distinguish the several remedies available (Macqueron, Flores Legum Scheltema, pp. 144–145, relies on this text unduly in his critique of Dig. 9.1.1.5); 1.15.3 denies a remedy where the victim has incited the animal, whether feram bestiam vel quamcumque aliam quadrupedem, against himself—impliedly providing a defence available against the actions summarised in both 1.15.1 and 1.15.2; and 1.15.4 gives an action, preserved also in the Digest title de extraordinaribus criminibus (Dig. 47.11.11) against “circlers” who cause loss as a result of the fright occasioned by their snake-handling.

Levy, Gesammelte Schriften, i.228, judges the author thus: “Wherever parallel texts enable us to check upon the character of the P.S. the picture is similar. The author, as a rule, did not mean to alter an opinion of Paulus. But his passion for brevity, coupled with a lack of faculty to think problems through, made him only too frequently distort the core of his master's analysis.”

11 Supra, note 4. Nor did Roman law admit measures of self-help comparable to the English distress damage feasant. See Dig. 9.2.39.1; Fliniaux, pp. 259–264.

12 Indeed, it seems that the owner might defend himself by alleging the culpa of his slave, though he would then be open to an actio damni iniuriae, Dig. 9.1.1.4.

13 Which declared the owner “free” where his ox, not already known to be a gorer (Ex. 21:29–30), killed a free man or woman. All the more so, it could be argued, should he be free if the injury were non-fatal.

14 Ex. 21:35, the result being a maximum award of half-damages. See Mishnah Baba Kamma 3:8; Tosefta Baba Kamma 3:1–2; Baraita Babli Baba Kamma 33a. There are indications in these sources that Rabbi Akiva (fl. early second century) argued in favour of full compensation.

15 Dig. 9.1.3.

16 , Lawson, Negligence in the Civil Law (Oxford 1950), p. 22Google Scholar note 3, has suggested that the praetor might occasionally have given an action on the analogy of the remedy for medical expenses and loss of earnings under the actio de pauperie. But there is no evidence of this. On the absence of a remedy for the negligent killing of a free man, see also Lawson (1940) 22 J.C.L.I.L. 142.

17 Dig. 21.1.40, 42.

18 E.g., Jackson, Essays, pp. 243–247. On Dig. 48.13.11.2 and Mishnah Baba Kamma 7:4 see now Daube (1977) 28 J.Jew.St. 80.

19 (1973) 46 Temple Law Quarterly 169–290.

20 c. a.d. 250. Cf. Cohen, B., Jewish and Roman Law (New York 1966), ii.579.Google Scholar

21 The rabbinic terminology for the distinction between Ex. 21.35, where the animal was not previously known to be a gorer (tam), and Ex. 21.36, where it was already an “attested” danger (mu'ad; from wehu'ad in the parallel Ex. 21;29).

22 Cf. the activities of xenokritai in administering Roman law in Arabia Petraea in the first half of the second century a.d., as appears from P. Eretz Israel 8:46–51; see Biscardi (1972) 19 R.I.D.A. 515–516; Festschrift für Erwin Seidl (Köln 1976), pp. 15–16, and literature there cited. A talmudic tradition, Baba Metzia 69a–b, claims that Rav Papa (4th cent. Babylonia) judged partnership cases between “Cuthaeans.”

23 Infra, note 26.

24 Discussed in my forthcoming Liability for Animals: An Historico-Structural Comparison. chap. 1.3.

25 Mishnah Baba Kamma 1:4, dating probably from early in the third century a.d. Second century authorities had already debated the status of the hyena and the snake, Tosefta Baba Kamma 1:4. The list is found in a slightly different context in Mishnah Sanhedrin 1:4, and probably originated there. But it is possible that its extension to civil liability was prompted by the greater knowledge of Roman law that resulted from the constitutio Antoniniana.

26 “Our Rabbis taught: The Government of Rome had long ago sent two commissioners to the Sages of Israel with a request to teach them the Torah. It was accordingly read to them once, twice and thrice. Before taking leave they made the following remark: ‘We have gone carefully through your Torah, and found it correct with the exception of this point, viz., your saying that if an ox of an Israelite gores an ox of a Canaanite there is no liability, whereas if the ox of a Canaanite gores the ox of an Israelite, whether Tam or Mu'ad, compensation has to be paid in full.… We will, however, not report this matter to our Government,’ Baraita Baba Kamma 38a; see Malamed, E. Z., Halachic Midrashim of the Tannaim in the Talmud Babli (Jerusalem 1943), 132Google Scholar no. 189, for other versions. Both this passage and that quoted above in the text appear as commentary on Mishnah Baba Kamma 4:3, the discriminatory rule here objected to by the commissioners. (See also the half-hearted attempt by Maimonides, Damage to Property 8:5, to justify this Mishnah in terms of conflict of law rules.) In another rabbinic tradition, Jerusalem Talmud Sanhedrin 1:4 (Krotoshin ed., 19b 19–24), Rabbi Johannan ben Zakkai is asked by the Roman governor to explain the Biblical rule requiring the death of the owner of a mu'ad which kills a free man or woman. The attributions in this group of traditions amount to a claim that the Romans took an interest in Jewish law in the last decades of the first century a.d. On rabbinical studies in Rome itself in the first two centuries, see Cassuto, U., Biblical and Oriental Studies (Jerusalem 1973), p. 291.Google Scholar

27 The one Roman text which may have been influenced by Rabbinic (as opposed to Biblical) law—and the possibility is remote—is Dig. 9.1.1.10, et ideo et si eum occidi, meum corpus esse, which introduces an issue hardly relevant to the title. Jewish law had earlier debated whether self-help was permissible against, inter alia, an homicidal bear: Mishnah Sanhedrin 1:4.

28 Dig. 9.1.1.4. Cf. Inst. IV.9.pr.

29 The Law of Obligations in the Later Roman Republic (Oxford 1967), pp. 280–281.Google Scholar He attempts, however, to explain them in terms of the classical law as evidence that the kicking or goring, “by themselves neutral acts,” occurred commota feritate, though conceding that Servius' pupil, Alfenus, did not insist on this requirement in Dig. 9.2.52.2. However this may be, it must be noted that the text has been interfered with, as is shown by the abrupt ending of the examples aut mulae propter nimiam ferociam (to which Mommsen proposed the addition of something like plaustrum everterinf); and that the equus calcitrosus and bos cornu petere solitus were especially favoured by the Byzantines, who selected the horse and the ox, but not the mules, as their prime examples of the action in the Institutes.

30 Cf. Lebigre, La Responsabilité pénale, p. 19 note 1, observing that bos cornu petere solitus looks astonishingly similar to the Laws of Eshnunna, § 54. Its similarity to the Biblical version of scienter was noted as long ago as Grotius, Annotationes ad Vetus Testamentum on Ex. 21:29, Opera Omnia Theologica (Amsterdam 1679), i.53. See also Card v. Case (1848) 136 E.R. 1022, where counsel observed: “The principle of a mischievous disposition in the animal, as a condition of the owner's liability, is recognised by Servius…” Indeed, the English civilian J. Cowell, Institutiones Juris Anglicani (Cambridge 1605), IV.8.1, had earlier attempted to read Ulpian's examples, two of them barely disguised, into his account of the English deodand, which he compared to the Roman noxal action: De brutis tamen animalibus & rebus etiam inanimatis, quorum violenta motione homo quisquam occiditur, quiddam noxalis actionis simile habemus. Si enim vel equus calcitrosus, vel taurus ferocior vel denique plaustrum a iumentis tractum hominem Occident, hoc quicquid est (nisi ad regem spectet) quasi sacrum sit: & nescio an consuetudine veteri comburl soleat: sed hodie Regi cedit.

31 For evidence of Justinianic awareness of Latin Bible translation (as if it were necessary), see Nov. 162, of a.d. 553; Cassuto, Biblical and Oriental Studies, p. 294.

32 Vulg.Ex. 21:29, Quod si bos cornupeta fuerit ab heri et nudiustertius et contestati sunt dominum ejus; Ex. 21:36, Sin autem sciebat quod bos cornupeta esset ab heri et nudiustertius. This is not, of course, to suggest that the theme of the goring ox was new to Roman law, or even that the categorisation of such a beast—for other purposes—as vicious was unknown. See Dig. 21.1.43.pr., Bovem qui cornu petit vitiosum esse plerique dicunt, of which Nicholas [1958] Acta Juridica 186–187, correctly observes that vitiosus refers to defects required to be declared by the aedflician edict on sale. Note, however, the interesting vitio bovis in Dig. 9.2.52.3.

33 1.15.2; supra, note 10.

34 Inst. IV.9.1.

35 The Animals Act 1971; the Guard Dogs Act 1975; the Dangerous Wild Animals Act 1976. Neither of the two latter are concerned with civil liability, though both dogs and animals which belong to a dangerous species constitute special categories under the 1971 Act. The 1975 Act seeks to protect the public by a combination of criminal sanctions and licensing requirements. The 1976 Act is not primarily concerned with protection of the public, but makes the holding of liability insurance a condition of the granting of a licence to keep the species to which the Act applies.

36 My thanks to Professor J. A. Ankum, Dr. A. F. Rodger, and Professor P. G. Stein for help and critical comment at various stages in the preparation of this paper. For the errors from which I have not allowed them to save me, they are of course blameless. I am grateful also to Professor J. M. Modrzejewski for making available to me proofs of his two articles in advance of publication.