CHAPTER I. PERS ONA — CAPUT — CORPUS — UNIVERSITAS

PERSONA

The word persona has a long and complicated history.

Legal Personality nowadays, Personnalitd Juridique, Juristische Persönlichkeit, is a highly technical term of jurisprudence. It means the capacity for legal rights and duties; and an entity capable of legal rights and duties is called a Legal Person. All human beings in civilised countries are Legal Persons. Where slavery exists, or existed, it is usual to deny Legal Personality to slaves, although they have not always been wholly without legal rights. But besides individual human beings there are other entities which have been treated by many legal systems as capable of rights and duties. The most familiar example is the Corporation Aggregate, a body of men joined together in a group which is recognised by the law as capable of rights and duties altogether distinct from the rights and the duties of its individual members. States and municipalities can be classed as Corporations Aggregate, together with voluntary associations, com­panies, and societies. Among other entities which have been regarded as Legal Persons may be mentioned the Corporations Sole of English law, charitable foundations, and the idols of present-day India. Each of these entities has been allowed by courts to bring actions; that is to say, individual men have been allowed to appear in court in the name, and as representatives, of the entities; and any entity, whatever its intrinsic nature, is properly styled a Legal Person if it is allowed by the courts to be a plaintiff or defendant.

Roman law certainly recognised some Legal Persons other than individual human beings; and the Latin word persona is sometimes used nowadays as a synonym of both Legal Person and Legal Personality. We find English judges saying ‘A corporation is a legal persona, just as much as an individual’,¹ and again ‘the idol...has a status as a separate persona ’ ·/ and it would be natural to suppose they are using a technical term of Roman law. It is the object of the following pages to find out whether this is true, whether the word persona was used by Roman

¹  See Sir F. Pollock, Essays in the Law, 1922, p. 155.

²               See Duff, ‘The Personality of an Idol’, in Cambridge Law Journal, J9²7» P· 43·

lawyers to mean either a Legal Person or Legal Personality, or, as a term of art, in any other technical sense. With this end in view we must first outline the earlier history of the word, and then discuss its use in connection with slaves, who were human beings but not Legal Persons, and with such other entities as were, or may have been, recognised as Legal Persons although they were not individual human beings.

The earliest traceable meaning of -persona¹· is a mask, such as Greek and Roman actors regularly wore on the stage.

Sometimes a man plays more than one part. Thus we read in the Digest (45. 3. 1.4): ‘Communis serous duorum seruorum personam sustinet.'^ He comes on the legal stage now in the character, wearing the mask, so to speak, of z^’s slave, now in that of 5’s, like Goldoni’s Truffaldino.

1       Cf. Schlossmann, Persona und TTp6wn-ov im Pecht und im christlichen Dogma, 1906, pp. 11-21.

2                                                                         Cic. Phil. 8. 10. 29.   ³ London, 1828.

⁴lang=EN-US> Cf.

We might say he appears in two capacities. Sometimes again the part played is not that of an individual man but of a personified abstract like Kpdro$ and Bia in the Prometheus Vinctus. This personification is defined by the rhetorician Rutilius Lupus (u. 6, ed. Halm, p. 15) ‘ripoCTco- iroiroiia. Hoc fit cum personas in rebus constituimus, quae sine personis sunt.... Nam humana figura produxerunt per­sonas, quae in veritate artis et voluntatis sunt, non personae'¹ Lastly, a man may impersonate, or represent, a community, and Cicero can say ‘NLagistratus gerit personam civitatis' (De Off. 1. 34. 124), and 'qui personam populi Romani sustinerent' (De domo, 52. 133).

Very often where persona is used without any apparent reference to the original meaning, it can nevertheless be explained by it. For example, lawyers often speak of one man’s bringing an action ex persona of another—the heir ex persona mortui, the tutor ex persona pupilli, the procurator ex persona domini. In all such cases, the underlying idea is that the man actually in court does not come as himself, but as ‘representing’ someone else. It is not far from ‘I, one Snout by name, present a wall’ to ‘A.B. suing as the legal representative of X.Y. deceased’.

The next development brings persona nearer to the ‘person’ of ordinary English usage. The transition may perhaps be seen in the familiar phrase ‘ dramatis personae'. This may originally have been a list of the masks that would be worn in the play; more probably it was meta­phorical from the first, and meant the parts to be played; but it was very natural that it should come to mean ‘the persons of the play’, ‘the characters’.^{[1] [2]} Thus persona may mean either a person whose character is portrayed on the stage,¹ or the actor who portrays it; and this leads³ to the meaning, perhaps the commonest in legal texts, para­phrased by Facciolati as ‘ ipse homo., quatenus hanc vel illam personam gerit', a man (or woman) in so far as he (or she) plays this or that part.

There is a very common variant of this use which is hard to analyse and often gives trouble to a translator. This is the use of persona with a genitive, apparently as a mere periphrasis—persona Titii meaning simply Titius (or possibly ‘Titius, bearing in mind his legal position’). Thus when Gaius says ‘ceteras vero liberorum personas si praeterierit testator' in one passage,⁴ and ‘ ceteri vero liberi sive masculini sexus sive jeminini una mancipatione exeunt

¹Cf. Plautus, Persa, 783; Schlossmann, pp. 15-17.

²               This development is discussed very fully by Schlossmann, pp. 11—31. His account is convincing, though he seems over-confident about details.

³ When the King says ‘Our Person’ or a subject ‘Your Majesty’ do

they mean ‘Me’ and ‘You’ or ‘Me (you), bearing in mind that I am (you are) King’? ⁴ G.

Enough has now been said on the general history, and we may turn to the specific question how far persona is used as a technical term corresponding to the modern Legal Person and Legal Personality. This may be divided under two headings: Do the Romans call slaves personae ? and Do they use the word of any Legal Persons other than human beings?

On the Personality of slaves there has been much con­fusion, because it has been assumed uncritically that persona means a Legal Person. We have already seen that one meaning of persona is ‘human being’, and slaves are certainly human beings; but it has been maintained that only those human beings are personae who are capable of legal rights and duties. This is suggested, though not stated, by Girard :^{[3] [4] [5]} ‘ L’esclave n’est pas une personne, mais une chose.’ It appears explicitly in Moyle’s Institutes.3 ‘What did the Romans mean by “persona”?...They did not regard all men as persons.... It is true that now and then, though very rarely, the word “persona” is applied to slaves (e.g. Bk i, 8 pr.); but the uniform language of legal authorities is the other way, and there can be no doubt that it is only per incuriam that occasionally a writer using this or other terms (such as caput, status) implying personality includes the slave as well as free persons within his view.... A persona is a man regarded as invested with legal rights, or as capable of acquiring them.’ This extreme view is quoted to show how dangerous it is to generalise hastily on this topic; for the facts altogether contradict the above statement. It was pointed out a century ago,¹ if not before, that the Roman lawyers ‘neither confined it’ (the term ‘-person'} ‘to human beings, considered as invested with rights; nor did they even restrict it to human beings, considered as subject to obligations. The meaning which they attached to the term, is the familiar or vulgar meaning. With them ¹¹ persona" denoted “homo"^ or any being which can be styled human? The truth of this has been finally proved by Professor Buckland, first in the Law Quarterly Review* and again in his Roman Law of Slavery·?» he quotes from legal writings no less than twenty-seven passages which speak of slaves as personae^ and makes it clear that this is the regular practice of lawyers from Julian and Gaius to Tribonian.

On the other hand there are a very few texts which show a tendency, beginning late in the third century, to use persona in a sense something like ‘capacity’, and to say that some men have either no persona or not the kind of persona that they need for the purpose in hand. One earlier text is quoted as denying persona to a slave, but it does not really do so. This is Digest 45. 3. 26 (Paul):⁴ ‘ Usus fructus sine persona esse non potest et idea servus hereditarius inutiliter usum fructum stipulatur.' This one might no doubt translate: ‘There can be no usufruct without a Legal Person, and therefore a slave who is part of an inheritance cannot validly stipulate for a usu­fruct.’ That looks as if the slave was unable to stipulate because he is not, or has not, a persona. But really, as Professor Buckland points out, it would be irrelevant to consider whether the slave is a Legal Person or not. The usufruct certainly cannot vest in him—'servile caput nullum ius habet’S—and the question really at issue is whether it can vest in the hereditas iacens and remain in suspense till there is a human being, a heres^ in whom it can vest.

¹ John Austin, Lectures on Jurisprudence (written before 1833), 3rd ed. 1869, vol. 1, pp. 360 f.                                                                                ’ 1901, p. 179.

3                            1908, p. 4.                                    ⁴ =Vatican Fragments, 55.     ⁵ D. 4. 5. 3. 1. We may translate either ‘ There can be no usufruct without a human being for it to vest in ’ or simply ‘ without some­body for it to vest in’. There seems to be a close parallel in D. 36. 1. 57. 1 (Papinian), which says that an inheri­tance cannot be immediately handed over to a slave who is set free only 'in tern-pus' (which must here mean after, not up to,class=a1> a certain date), ‘ quando persona non est, cut restitui potest' This might obviously mean ‘ since there is no Legal Person to whom it can be transferred’, but it can equally well be rendered ‘since there is nobody to whom it can be transferred’. In both passages the word is ambiguous and must be interpreted in the light of the many texts where it is not. Since there are no other texts as old as these two where persona seems to mean Legal Person, and many where it is used of slaves, we must conclude that the word here too means ‘human being’ and not ‘Legal Person’.

The evidence is more nicely balanced in a text not mentioned in this connection by Professor Buckland, namely D. 41. 1. 61 (Hermogenian): 'Hereditas in multis partibus iuris pro domino habetur adeoque¹ hereditati quoque ut domino per servum hereditarium adquiritur. In his sane, in quibus factum personae operaeve substantia desideratur, nihil hereditati quaeriper servum potest, ac propterea quamvis servus hereditarius heres institui possit, tamen quia adire iubentis domini persona desideratur, heres exspectandus est. (§ 1) Usus fructus, qui sine persona constitui non potest, hereditati per servum non adquiritur.' The last sentence seems to be merely quotation of Paul, but 'factum personae' is a curious expression. ‘An act of a Legal Person’ would make sense; ‘ an act of a human being ’ does not. Obviously a slave’s stipulation is ‘an act of a human being’; and a servus hereditarius can validly stipulate? It seems best, though not very natural, to translate factum personae ‘an act of the person who is to be entitled’. But, since the text is from Hermogenian, who wrote in the fourth or ¹ ideojue F’.    ¹ D. 41. 1. 33. 2. very late third century,¹ it is conceivable that we have here an adumbration of the modern technical meaning.

Similar adumbrations may be found in the Codes of Theodosius and Justinian. We hear as early as a.d. 294’ of ‘legitimae administrations persona' in connection with a tutor who gave no security: ^lea, quae ab eo gesta sunt, qui legitimae administrations personam sustinere non potuit, ipso iure irrita sunt'. Here sustinere suggests that the stage metaphor is still alive; the unqualified tutor, like the incompetent actor, cannot play the part he undertakes. We may translate ‘who cannot lawfully act as admini­strator’ or ‘perform the part of an administrator’. The chief thing to note about this text is that persona cannot possibly mean Legal Person or Personality. We will now consider a series of texts where that meaning gradually becomes possible and ultimately perhaps probable.

Let us take first C. 4. 46. 3,3 a.d. 337, dealing with the public sale of property confiscated for non-payment of taxes: 'sin autem minoris forte persona fuerit inserta (observe that a minor is, or has, a persona), necesse sit legitimae defensionis adesse venditioni personam', a lawful guardian must attend the sale. This is very like C. 2. 40. 4, and legitimae defensionis persona must mean almost the same as qui legitimae administrations personam sustinere potest, i.e. a duly qualified and appointed curator. Defensio is the röle, persona the player. Persona does not here mean a Legal Person; but the phrase legitimae defensionis persona does mean, not indeed someone capable of rights (rechtsfähig)— a minor is that—but someone capable of taking part in legal transactions (handlungsfähig); and the step between these is short.

We come now to the ten passages in the Codes that

¹ P. Krüger, Geschichte der Quellen und Litteratur des röm. 'Rechts, 2nd ed. 1912, p. 254; Kalb, Roms Juristen, 1890, p. 144.

^z C. 2. 40. 4. The lex gemina, C. 5. 42. 3, is dated 287; but Krüger, adloc., says that is wrong and 294 right. The clause qui.. .potuit is not in the lex gemina, and may well be interpolated.

³               =Th. 11. 9. 2 (corrupt at this point).

IO contain the words persona legitima with various meanings. One may be dismissed as irrelevant to our problem, namely C. 6. 58. 14 (a.d. 531), where legitimas personas means agnates—‘ legitimas personas, id est per virilem sexum descendentes'. In six of the others legitima persona means a person (i.e. a man or woman) legally qualified to do or be the thing under discussion.¹

Thus in C. 6. 17. 2 (=C. Th. 4. 3. 1), a.d. 393, ‘ Carbonianum edictum sub personis legitimis indubitato matri­monio...defertur'., personae legitimae must mean persons entitled by age, ius connubii, and so on, to contract a civil marriage, iustae nuptiae, entitling their children to benefit by the Roman law of succession. A slave would not be a persona legitima in this sense; but neither would a pere­grine or a Latin, though they were certainly Legal Persons.

C. Th. 3. 1. 7 (a.d. 396) runs: ‘Semel inter personas legitimas initus empti contractus et venditi ob minorem ad- numeratam pretii quantitatem nequeat infirmari' Here the Interpretatio gives ‘ duas quascumque personas', showing how little importance the interpreter attached to the phrase legitima persona. Those words here mean on the face of them ‘persons qualified to make a valid contract’, which would include slaves and exclude lunatics and perhaps those barred by special circumstances, such as a tutor contracting with his ward or an official buying land in his province. But in view of other texts we may guess that legitimus here means primarily ‘of full age’; if one of the parties was a minor there was always a risk of the sale’s being avoided.

This explanation of the word is given by the Interpretatio of C. Th. 2. 4. 4 (a.d. 385). The text is concerned with the course of a normal trial and the effect of imperial rescripts, and says that at a certain stage 'lis exordium auspicatur inter iustas videlicet legitimasquepersonas', parties capable of bringing and defending actions; but the Inter­pretatio is much more definite: 'si tamen litigantumperfectae

¹ The very expression ‘persona legitima' indicates that persona itself has no ‘legal’ connotation. So does our use of ‘legal persona'.

■probantur aetates', ‘if the litigants are shown to be of full age’. This certainly fits quite well; if one party was an unauthorised pupillus the action would be a nullity.¹

The same idea seems to underlie C. 3. 6. 3 (=C. Th. 4. 22. 6), of a.d. 414. This is one of the three laws in Justinian’s title legitimam personam in iudiciis habent vel non, and reads as follows: 'Momentariae possessionis actio (momenti actio C. Th.) exerceri potest per quamcumque personam, sub colore autem adipiscendae possessionis obrepticia petitio alteri obesse non debet, maxime cum absque conventione personae legitimae initiatum iurgium videatur, nihil autem opituletur conventio circa minorem habita, cum id rectius circa curatorem debuerit custodiri' By itself, 'absque conventione personae legitimae' is not clear, and Schlossmann (p. 71) doubts whether the genitive is objective or subjective, whether the persona who must be legitima is the plaintiff or the defendant; but the last sentence makes it reasonably certain that persona legitima means a defendant who is either of an age to defend himself or lawfully qualified to defend the minor whose right is disputed.

Similar to these texts is a passage of the Autun Gaius^{[6] [7] [8]} quoted by Schlossmann,3 which indicates that when it was written legitima persona was already familiar in the sense of ‘someone qualified to conduct a lawsuit’: ‘ Etideo] tractamus, quae sint legitimae personae, nam permittitur his qui haben\t litem, ut pro~\prio nomine consistant ut legitimae personae. Sed non solum per nos\met ipsos pos\sumus litigare, sed etiam per alias, non per quascumque, sed per certas person[as, scilicet per\ cognitores aut per procuratores aut per tutores aut per curatores, qui pro al\iis ag\entes habent legitimam personam.' Here the class of legitimae personae includes not only tutors and curators but also cognitores andprocuratores, whose principals would be equally legitimae personae if they came to court themselves.^[9]

The last of the six Code passages is C. 8. 5. 1. 1 (=C. Th. 4. 22. 1), of a.d. 326. This law allows anyone, slave or free, to sue in the name of an absent owner for return of possession wrongfully taken; and adds that no such litigation is to prejudice in any way the rights of the absent owner: 'omnia quae supererunt ad disceptationem litigii immutilata permaneant: iudicio servatoiustis legitimisque personis, cum valde suflflciat possessionem tenentibus absentium nomine contra praesentium violentiam subveniri.' Here iustae legitimaeque personae clearly means people entitled not only to bring an action² but to win it—the true owners of the property.

In all these texts legitima persona means someone who has some particular right or capacity: in none does it mean one capable merely of rights or liabilities in the abstract. It is opposed to minors, and to those who falsely pretend to be tutors or owners, but never to slaves.

Next comes a group which it is hard to discuss accurately and without pedantry. Hitherto we have been concerned with texts where homo could have been substituted for ■persona without much change of sense. In the following texts this would not be possible. Here persona is used in senses derived more directly from the meanings ‘mask’ and ‘rdle’. But we have always to remember that persona Titii is often indistinguishable in meaning from homo Titius or ipse Titius; and it seems that legitimam personam habere, gerere, sustinere may also be indistinguishable from legitima persona esse.

Thus in the rubric of C. 3. 6, * §>ui legitimam personam in iudiciis habent vel non’, we may translate ‘who play a lawful part ’ or ‘ can lawfully play a part ’; but the meaning is exactly the same as if the words were ‘ Qyae personae in iudiciis legitimae sunt vel non, ‘what persons are lawfully qualified to be litigants’. The title contains only three laws. The third has been discussed, and the others are both concerned with the bringing of actions by or against minors with or without the authority of their tutors or curators.

So in C. Th. 10. 10. 27. § 4 (a.d. 418, misdated 415) ‘ statuto tempore legitimam personam sui probaturus adveniat' clearly means the same as 'legitimam personam se habere' or 'se esse probaturus', that he is playing a part for which he is lawfully qualified, or that he is lawfully qualified for the part he is playing. The nature of the part is not per­fectly clear, but it seems to be the part of owner of the property in question. It would certainly not be enough to prove he was a free man, a ‘Legal Person’.^[10]

In the last remaining passage of the Codes where we find the words ‘persona legitima' the precise meaning is ob­scure, but the general sense is similar. This is C. 6. 60. 1. 1 (=C. Th. 8. 18. i. § 3), of a.d. 315 (misdated 319), dealing with bona materna. It says that patresfamiliarum 'omnem debent tuendae rei diligentiam adhibere et quod iure filiis debetur.. .poscere...et litem infierentibus resistere atque ita omnia agere, tamquam solidum perfectumque dominium et personam gerant legitimam'. The paterfamilias is not owner of the bona materna, but he is to act in every way as if he was owner, except that he cannot validly alienate. Schloss­mann (p. 67) says: ‘er soli dominium gerere, d(as) h(eisst) nicht etwa: das Eigentumsrecht haben... der Kaiser will sagen, der Vater solle nur im Prozess wie der Herr der Sache auftreten dürfen...personam legitimam gerere besagt pleonastisch dasselbe, er solle von Rechts wegen eine Rolle haben, d. h. als Vertreter des Kindes im Prozess aufzutreten befugt sein’. This seems to ignore tamquam·, the text does not say fathers are to dominium gerere, but to act as if they dominium gerant·, and we may render ‘as if they were full owners and therefore had a lawful part to play in litigation’, ‘and were lawfully qualified to bring and defend actions’.

Two other relevant texts are C. Th. 8. 18. 3 (a.d. 334), 'maiorem sibi in.rebus filiorum vindicare personam', which evidently means ‘play a greater part, or röle’, and C. Th. 3. 17. i (a.d. 319; = C. 5. 34. 11), 'In universis litibus placet non prius puberem iustam habere personam, nisi' etc., ‘ cannot play a lawful part ’, ‘ is not a legally qualified person ’.

We are now ready to consider the two passages where slaves are actually said not to have a persona. These are Nov. Theod. 17. i. 2 (a.d. 439) and Cassiodorus, Variae, 6. 8. 2 (ed. Mommsen, Monumenta Germaniae Historica, Auctorum Antiquissimorum, tom. xii, 1894, p. 181, 1. 31; about a.d. 527). The Theodosian passage runs: 'Servos namque nec ab initio, quasi nec personam habentes, in iudicium admitti iubemus et, si hoc ausi fuerint, continuo eos flammis tradi vel bestiis.' Here we are clearly concerned with capacity to appear in court, not with capacity for rights, Prozessfähigkeit not Rechtsfähigkeit. But personam only

IS conveys this meaning in virtue of the context, and we can translate ‘ as having no part to play on the stage of the law­court’. The text is clearly parallel to those containing the phrase persona legitima, and neither that phrase nor the word persona here has any technical meaning at all.

This is equally clear in the Cassiodorus passage, part of the Formula Comitivae Privatorum. The Comes originally controlled only the slaves of the imperial domains, but assumed a wider jurisdiction, ‘ne dignitas Latialis causam tantum modo videretur habere cum famulis, sed actibus urbanis tunc se felicius occupavit, postquam agrestium causas decenter amisit, Quid enim prius facerent inter servos iura publica, qui personam legibus non habebant? Non ibi advocatus aderat, non se partes sollemni actione pulsabant'; etc. It seems impossible to doubt that personam legibus non habebant is exactly equivalent to personam legitimam non habebant, and means they had no lawful rble, no part given them by the law in the drama of litigation.

If we are right in explaining these two passages in this way, by reference to the texts which contain the phrase persona legitima, it may perhaps be taken as proved that Roman lawyers did not use the word persona as a technical term meaning either Legal Personality or a Legal Person.^[XI] A free man is never said to be, or to have, a persona, in contrast to a slave; and where slaves are said not to have a persona, this seems to be untechnical.

There are however three troublesome texts in the Greek paraphrase² of Justinian’s Institutes traditionally attributed to Theophilus, part-author of the Institutes themselves. All three describe slaves as Απρόσωποι, having no πρόσ- ωπον; and πρόσωπον is the normal Greek equivalent of persona. They are as follows.

Inst. 2. 14. 2 (Ferrini, vol. 1, p. 178): ών τούς δεσπότας¹ δυνάμεθα γράφειν κληρονόμους τούτων καί τον οίκέτην. ό γάρ οίκέτης άπρόσωπος ών έκ του οικείου χαρακτηρίζεται δεσπότου.

Inst. 3.17· Ρ^Γ· (Ferrini, vol. π, ρ. 328): οί οίκέται Απρόσω­ποι δντες έκ των προσώπων των οίκείων δεσποτών χαρακτηρί­ζονται, καί έκεϊθεν έξουσι την κατάληψιν πότερον έπερωταν δύνανται ή ού δύνανται.

Inst. 3. 29. 3 (Ferrini, vol. 11, ρ. 380): άλλα δει προς τοϋτο λέγειν ότι ποιεί novationa ού μόνον τό τίκτεσθαι φυσικήν ένοχήν, άλλα καί τό ύπεΐναι πρόσωπον · Απρόσωπος δέ παρά τοϊς νόμοις ό δούλος.

It cannot be denied that in all these passages Απρόσωπος can plausibly be translated as ‘without Legal Personality’. It is perfectly true that slaves have no Legal Personality of their own and therefore take Legal Personality from their owners. And it is possible that we have here a technical terminology which had grown up at Berytus or Byzantium and does not happen to have come to light in any earlier work. It is even possible that Theophilus himself, or whoever wrote the Paraphrase, felt the need of a word to express the incapacity of slaves for rights and duties, and pressed into service the very rare word Απρόσ­ωπος. But it does not seem very likely. Persona had been used for centuries without acquiring a technical meaning, and to a Byzantine lawyer persona and πρόσωπον seem to have been interchangeable synonyms.² The poets tells us that Justinian’s Pandects only make precise

What simply sparkled in men’s eyes before,

¹ This seems corrupt, των δεσποτών oOs δύναμαι gives sense, and is the reading of one MS.

¹ Theophilus himself has (1. 3. pr.) ή τοίνυν των προσώπων άκρα διαίρεσίς ίστιν αύτη· ότι των άνθρώπων οί μέν είσιν έλεύθεροι, οΐ δέ δούλοι. See Schlossmann, ρ. 58.

³ Or rather Count Guido Franceschini; Browning, The Ring and the Book, v, 1781. Cf. Maitland, Brocton's Notebook, vol. 1, p. 9.

17 but it is hard to believe that Theophilus’ Paraphrase made precise what in the Corpus Juris had been vague and nebulous.

The translation of απρόσωπο? in Theophilus as ‘without Legal Personality’ is plausible, and certainly possible, but not necessary. A less technical meaning is suggested by D. 28. 8. 1. pr.: ‘Si servus fuerit heres institutus, utique non ipsi praestituimus tempus ad deliberandum, sed ei cuius servus est, quia pro nullo isti habentur, apud praetorem'¹

‘Pro nullo habentur' is not easy to translate. Perhaps we might say ‘are ignored’ or ‘are not counted as any­body’. The point seems to be not that they cannot have rights, but that their wishes are not taken into account. What they want is of no consequence; their master’s choice is everything.

So in the first two Theophilus passages Απρόσωπο? seems to mean, not incapable of rights, but without any legal character of their own, positive or negative. We have seen above that persona commonly means the part played by a man in the drama of the law; these texts tell us that slaves are not cast for any part, and whether they can be instituted heirs, or stipulate, depends on the rble played by their masters, whether as citizens, peregrines, or lunatics. They are in themselves colourless, like the glass jars that a chemist fills with red and green and purple liquids, and take their colour from their masters. To say that a slave cannot himself be the subject of rights would be true; but it would not be very relevant.

The third passage is more difficult. A slave’s promise does not novate, although it creates a natural obligation, because for novation δει ύπεϊναι πρόσωπον Απρόσωπο? δέ παρά τοϊ? νόμοι? ό δοϋλο?. It is tempting to translate ‘there must be a Legal Person; and a slave is not a Legal Person’. But the temptation is lessened by looking at the corresponding words of Gaius and Justinian. Gaius

’ Cf. also D. 50. 17. 32: 'Oyod attinet ad ius civile, servi pro nullis habentur I

l8 says (3. 176) 'proinde adhuc obligatus tenetur, ac si postea a nullo stipulatus fuissem'; and Justinian (3. 29. 3) only changes tenetur to manet and fuissem to fuisset. Buckland¹ comments: ‘ This is not the real reason, for the statement is incorrect: it creates a naturalis obligatio, apart from the edictal rules. In fact the statement, though it has been used for far-reaching conclusions, is obscure.’ It is not surprising that the Paraphrase should give difficulty. Gaius certainly seems to say that the first debtor remains bound as if the second stipulation had been made with Nobody, and to imply that for novation you must stipulate with Somebody. On that unhelpful remark it is a reason­able comment to say that for this purpose the law regards a slave as Nobody, especially as we are told elsewhere that slaves pro nullo habentur·, and that seems to be all Theophilus is trying to say. It is very doubtful if he really meant, as appears on the face of it, to make a general statement; and very unlikely that in using the word ccrrpoCTcoiros he had any very definite meaning in his mind at all. Any theory based upon the word orrpocrcoTros must be built on sand.^{[12] [13]}

The texts so far dealt with give scanty and treacherous support to the view that persona or ττρόσωπον can be used technically to mean a Legal Person, a Right-and-duty- bearing Entity. We must now consider whether the word persona is ever used of Legal Persons other than individual human beings. Most of the relevant texts fall into two groups, one concerned with the hereditas iacens, the other with towns and their landed property.

The word persona occurs in several passages of the Digest and the Institutes in connection with the hereditas iacens, the inheritance on which no heir has yet entered: and partly for that reason, the hereditas has been regarded by some civilians as itself a Legal Person. That view will be discussed later;¹ the present question is not whether the hereditas was a Legal Person, but whether it was called a persona·, and the answer is in the negative. A hereditas is said personae vice fungi,² to represent, take the place of, a human being; the law ignores, for some purposes, the fact that at the moment no human being is owner, debtor or creditor. The dead man has ceased to play his part, he has ceased to wear his mask; soon the heir will put it on, step into his shoes, as we say, and continue the play; but meanwhile the hereditas iacens provides a peg on which the mask can hang. The question our texts are chiefly occupied with is whether the hereditas represents the dead alle Rechtssatze, die Rechte verleihen und Pflichten auferlegen,... finden auf Skiaven keine Anwendung’—rules of law which give rights or impose duties have no application to slaves. That is perfectly true, but very different from saying that slaves do not exist. It is equally true of animals and tables and chairs; but we should be surprised to find a lawyer saying ‘horses do not exist’. It seems clear that by ‘do not exist’ Schlossmann means ‘are not capable of rights and duties’, ‘are not Legal Persons’; which is the rendering of Απρόσωπος that he most strenuously opposes. On the other hand it is quite possible that theological usage may have made easier the use of Απρόσωπος παρά τοΐς νόμοις as an equivalent of pro nullo.

¹ Chapter vii.

¹ D. 46. 1. 22 (? interpolated. See p. 20, infra)·, D. 41. 3. 15. pr.: ‘nam hereditatem in quibusdam vice personae fungi receptum est'. The pr. is. very corrupt, but this sentence seems to be sound. man or the heir, whether the acts of a servus hereditarius, for instance, are to be construed as if the dead man were still alive, or as if the heir had already entered. Generally speaking, it is the legal character of the dead man that determines the rules applicable to his estate. Consequently we find 'hereditas personae defuncti, qui eam reliquit, vice fungitur',¹ 'nondum enim adita hereditas personae vicem sustinet, non heredis futuri, sed defuncti'',² 'sed hereditas in plerisque personae defuncti vicem sustinet',3 'hereditas enim non heredis personam, sed defuncti, sustinet'.⁴ But this rule is not without exception, and we also find 'transit ad heredem, cuius personam interim hereditas sustinet'.5 None of these passages says that a hereditas is a persona, only that it represents or is treated like one.

One of the texts quoted takes us beyond the hereditas. This is D. ₄6. i. 22 (Florentinus): 'hereditas personae vice fungitur, sicuti municipium et decuria et societas'. Interpola­tion is strongly suspected;⁶ but whether the passage was written by Florentinus or Tribonian, it does not say that any of these bodies are style='font-style:italic'>personae. It was certainly true in the time of Florentinus that some things, such as the making of contracts, normally done by and on behalf of personae, individual human beings, could be done on behalf of a hereditas tacens or a municipium f and the text says no more than that.

Another passage quoted⁸ to show that persona can be used of a collective person is D. 4. 2. 9. 1, where Ulpian has the words: 'sive singularis sit persona, quae metum intulit, vel populus vel curia vel collegium vel corpus'. This

’ D. 30. 116. 3 (Florentinus).                       ² Inst. 2. 14. 2.

³   Inst. 3. 17. pr.

⁴   D. 41. 1. 34 (Ulpian). Cf. D. 41. 1.33.2, where a servus hereditarius

'ex persona defuncti vires assumit’.                5 D. 46. 2. 24 (Pomponius).

6       See Schnorr von Carolsfeld, Geschichte der juristischen Person, vol. 1, 1933, P· $2, for references to Rabel and Scaduto.

7       That it was also true of some second-century societates is maintained below, p. 144; for decuriae, see pp. 32 and 101, infra.

8       Saleilles, Personnalitdjuridijue, 2nd ed., 1922, p. 77. is interesting, as it shows he thought it possible, or at least conceivable, for a group "meturn inferre'J But it is not very good evidence that he would have called a group a persona. It is true that there is a shade of difference between singularis and singulus., and that the expression singularis persona may suggest the possibility of a persona non singularis. But this argument cannot be pressed. If Ulpian had said singularis homo no one would have been surprised, and the meaning would have been exactly the same, ‘an individual’. The implication is not that a corpus is a persona, but that a corpus and a persona are two entities alike in their power of creating fear but different in that a face="Times New Roman">persona is singularis, a corpus complex and composed of several personae acting as a unit. If they are not acting as a unit, they are singulares personae \ but whatever they do, ten men will never make one persona or eleven personae. 'Singularis' may imply the idea of a corporation, a uni­versitas, which was familiar to Ulpian;² but not that a universitas could be called a persona. A very similar text is C. 2. 58. 2. 5, where Justinian says *sive pro una persona quis litem movere voluerit sive pro aliquo corpore vel vico vel alia universitate'. Una persona is clearly contrasted not with duae singulares personae but with the universitas, which is not una persona though it may be ‘une personne’.

Two other texts are quoted from the Digest for singularis persona. D. 35. 1. 56 is concerned with conditional legacies. If a farm is left to A, conditionally on his paying ten, he cannot get part by paying part. But if it is left to A and B, each can get his share by paying his share. 'In eo vero, quod uni sub condicione legatum est, scindi ex accidenti condicio non debet, et omnis numerus eorum, qui in locum eius substituuntur, pro singulari persona est habendus.’ Here there is clearly no suggestion of a corporate or collective person. Pro singula or pro uno would give the same sense. The rule to be applied is that normally applic-

¹                      See p. 91, infra, and Schnorr von Carolsfeld, p. 339.

²                       Cf. p. 37, infra.

able to a single legatee, not that appropriate to a joint legacy. D. 50.16. 195. i (Ulpian) is obscure: ‘ ad personas autem refertur familiae significatio ita cum de patrono et liberto loquitur lex:¹ “ex ea familia”, inquit, “in eam fami­liam”: et hic de singularibus personis legem loqui constat'. This may perhaps refer to the well-known fact that a Roman familia can consist of a single individual and a citizen sui iuris is none the less a paterfamilias because he has no wife or child. There is clearly no suggestion of any corporation or universitas, and whatever the truth may be about family ownership in primitive times,^{[14] [15] [16]} Ulpian would certainly not have said that a Roman family, husband, wife, sons, daughters, and slaves, composed a persona.

Another text which may be quoted in this connection is C. I. 3. 48. 7, part of which SaleillesS paraphrases: ‘Tout ce qui est ldgu£ ä un monast&re, ä une Eglise, appartient ä l’Eglise et d’une fa^on g^nirale ä la venerabilis domus, en tant que personne certaine, persona certa' This looks as if persona was applied here to the Church and to all charitable institutions. But the Latin is 'sin autem in personam certam vel in certam venerabilem domum respexit, ei tantummodo hereditatem vel legatum competere sancimus'. This cannot mean what Saleilles says, and the most natural way of taking the words is to make them contrast gifts to a domus with gifts to a bishop, an oeconomus, or a xeno­dochus, who are all mentioned in this constitution. This gives excellent sense, and allows persona to bear its regular sense of ‘a man’.

There are, however, at least two passages in the litera­ture of land-surveying where the word persona is used of a town. Frontinus, who died about a.d. 105, in his work De Controversiis Agrorum? contrasts two forms of words used in map-making: ‘ silvae et pascua publica Augustinorum', and 'silva et pascua', or ‘fundus Septicianus, Coloniae Augustae Concordiae'; and of the second form he says: ‘ Haec inscribtio videtur ad personam coloniae ipsius pertinere' The distinction is obscure, though the language is some­thing like that in Gaius 2. 11: ‘ipsius enim universitatis esse creduntur'. But there can be no doubt that for Frontinus a colonia had apersona, and ad personam coloniae was another way, ad personam coloniae ipsius a more emphatic way, of expressing ad ipsam coloniam. The second passage, probably also from Frontinus, is preserved in two slightly different forms, with the commentary attributed to Agennius Urbicus, a Christian schoolmaster of the fourth or fifth century. On p. 80 of Lachmann’s edition,^{[17] [18]} we read: ‘Nunc ut ad publicas personas respiciamus, coloniae quoque loca quaedam habent adsignata in alienis finibus, quae loca solemus praefecturas appellare. (Comment: harum prae­fecturarum proprietas manifeste ad colonos pertinet, non ad eos quorum fines sunt deminuti.?^ This reappears in the cento called Commentum de Controversiis* as: ‘ Quaedam loca feruntur ad personas publicas attinere, nam personae publicae etiam coloniae appellantur, quae habent assignata in alienis finibus quaedam loca, quae solemus praefecturas appellare, harum praefecturarum proprietates manifeste ad colonos per­tinent, non ad eos quorum fines sunt deminuti' Here again Frontinus seems rightly interpreted as identifying coloniae with personae publicae·, and it does not seem to fit at all well into the general usage of the word persona. We may perhaps conjecture that the surveyors, who were used to

24 diagrams and symbols, were particularly open to the influence of pictorial personifications.¹ Towns were repre­sented on coins in human shape long before the Christian era, and Roman religion gave a deeper significance to such personification than we attach to it now: Father Tiber was a more real person than Father Thames, and Roma Dea than Britannia. One thing only is clear: that neither Frontinus nor Agennius is using persona as a legal term of art. Like Theophilus, they suggest questions, but do not justify any answer.

Finally, Schnorr von Carolsfeld² calls attention to the use in Christian writers of such expressions as ‘ in persona Christianorum'^ and ¹ ad personam populi huius',⁴ where there is something like personification of the community, or at least some emphasis on its being a community, not a haphazard plurality of individuals; and5 to the influence of the common Hebrew word O'JS, panim, which is translated by irpocrooTrov, persona, and ‘face’, and may be followed by the genitive of a man, a body of men, an animal, an inanimate object, or an abstract conception.⁶

We have now discussed all those uses of the vior&persona which seemed likely to throw light on the development of Legal Personality, and the light gained is extremely faint. Some texts? suggest that persona can be used to mean the capacity for legal rights, or for some particular legal rights. Others⁸ show a consciousness that the law does sometimes proceed as if there were a human being for rights to vest in, when in fact there is none; or9 a willingness to ‘per­sonify’ a community, to use of a town language literally appropriate only to an individual. But nowhere is persona truly a technical term. Nowhere does it seem at all likely

¹ Cf. Schnorr von Carolsfeld, pp. 5 5 f.; Schlossmann, p. 126.

’ Op. cit. p. 54.

³  Lactantius, De mortibus persecutorum, 48. 7.

⁴  Hilarius, in Matth. 12. 22.                 5 Schnorr von Carolsfeld, p. 56.

⁶ For numerous references see Schlossmann, pp. $4f.

1 See pp. 7—18, supra.                      ⁸ See pp. 19, 20, supra.

9 See pp. 22—24, supra.

*5 that in using the word any writer had present to his mind the ideas called up in that of a modern lawyer by the phrases Legal Person, Legal Personality, Personne Juridique, Rechtsfähigkeit. Strange as it may seem, the word -persona plays no part in the early history of Personality; and we must turn our attention to the less obvious but more fruitful study of caput, corpus and universitas.

CAPUT

A word which, in some contexts, corresponds more closely than persona to the modern Legal Personality is caput. From the sense of‘head’, it is extended Wk&persona, ‘mask’ (and still more like irpdaxoirov, ‘face’), to mean both ‘a man’ and ‘the position of a man’. Plautus uses it of slaves as well as free men, especially in the periphrasis tuum caput for Zw;¹ but also apparently in the sense of libertas? So lawyers sometimes speak of a slave’s caput, sometimes say he has none. In the familiar rule ‘ noxa caput sequitur'the caput followed is clearly that of the delinquent, who is normally a slave, and must mean his natural person, his body and soul, which persist through all changes in his legal position. So in D. 4. 5. 3. 1 Paul says that manu­mission involves no capitis minutio because 'servile caput nullum ius habet'. Evidently a slave is or has a caput. But the Institutes⁴ explain the same rule by saying that the slave 'nullum caput habuit'; we are a long way here from the literal meaning, and must suppose that the sense is something like ‘Legal Personality’. Some such sense is required by the phrase capitis minutio itself, which goes back to the Republic.5 The thing that is here diminished

¹               E.g. Miles Gloriosus, 326.

²               E.g. Rudens, 929: ‘pollicitabor pro capite argentum ut sim liber'·, Mostellaria, 211.

³                                                                                                                                                                                   lang=EN-US>D. 47. 2. 18; P. 2. 31. 8; G. 4. 77.        ⁴ Inst. 1. 16. 4.

⁵               Cicero, Topica, 4.18. Horace assumes its familiarity in the Regulus Ode, 3. 5, 11. 41-3: Fertur pudicae coniugis osculum Parvosque natos ut capitis minor Ab se removisse.

or destroyed is clearly position in the eye of the law, civil capacity, Legal Personality. So too iudicium capitale is not necessarily one that may end in decapitation; enslavement and deportation are also forms of capital punishment.¹ But we need not go further into the thorny problems of capitis minutio, its meaning and history; for whoever had or had not caput, however it was or was not lost, it is never attributed to any being other than a natural man, to any town³ or college, in short to any Juristic Person.3

CORPUS

Corpus, on the other hand, meaning a ‘ body ’, can obviously be used appropriately of a ‘corporate body’ and is very commonly so used. Any set of men who are for any pur­pose regarded as a group may be called a corpus, as we are told in D. 41. 3. 30. pr. (Pomponius, L. xxx ad Sab.'): ‘ tertium (sc. genus corp or urn), quod ex distantibus constat, ut corpora plura [wo»]⁴ soluta, sed uni nomini subiecta, veluti populus, legio, grex'. Schnorr von Carolsfeld5 says indeed that corpus here is ‘bestimmt gearteter Formbegriff; der Inhaltsbegriff braucht sich mit diesem in keiner Weise zu decken’. If this means that Pomponius’ definition covers groups or classes or categories not normally thought of as groups or likely to be called corpora, it is true. ‘ Men named John’, ‘all mountains higher than Mont Blanc’ are uni nomini subiecta, but are not groups in the same sense as a family or a mountain range. But Pomponius’ examples are all groups of the second kind, and it seems fairly clear

¹ D. 48. 1. 2.

³               A town may indeed be called caput regionis, the capital of a country; but that is different.

³               On caput, see Clark, History of Roman Private Law, vol. 11, 1914, pp. 457—9; and on status, ibid. pp. 483—91, and Allen in Law Quarterly Review, 1930, p. 277. This word often occurs in connection with capitis minutio, but is generally recognised to be quite untechnical, meaning little more than ‘position’.

⁴ non del. Bemays.                                       ⁵ Op. cit. p. 147.

that by corpus quod ex distantibus constat he means, not a category of individuals whose only unity lies in a common name, but a complex of individuals connected by some intrinsic relation, though not spatially conjoined.¹ Uni nomini subiecta is perhaps misleading, but the Romans, like the English and unlike the Germans, were bad at defining, though quite good at giving examples.

The question then arises whether we can find a second narrower and more technical sense of the word. Three such have been suggested: that corpus is properly a cor­porate body, a body recognised by law as a Legal Person that corpus is properly a body recognised by law as useful to the State and therefore given privileges not enjoyed by other bodies ;3 and that corpus can be used to mean Legal Personality itself, Rechtsfähigkeit, the quality which distinguishes corporations from what English lawyers call unincorporate bodies. This third sense, suggested by Mitteis,⁴ is discussed below,5 in connection with the rules about collegia, and the conclusion is reached that in one fragment of the Digest,⁶ though nowhere else in Latin literature,⁷ corpus is used, and used twice over, to mean Corporate Character, a concept which, though certainly known to the Byzantines, and almost certainly to early

2       Tn engerer, eigentlich technischer Bedeutung steht corpus für die rechtlich als öffentlicher Organismus anerkannte, mit den Rechten einer juristischen Person beliehene Genossenschaft’, Kornemann in Pauly- Wissowa’s Realencyclopädie, s.v. collegium, p. 380; ‘für den Verband mit juristischer Persönlichkeit’, Leonhard, ibid. s.v. corpus, p. 1645.

3       Waltzing, ftude Historique sur les Corporations Professionnelles chess les Romains, vol. 11, 1896, p. 140.

4       Mitteis, Römisches Privatrecht, 1908, pp. 399 ff.

5       See pp. 141-9, infra.

6       D. 3. 4. 1. pr. and § 1.

7       Mitteis (p. 394, n. 13) cites Livy, 26. 16. There is to be at the dis­graced Capua 'corpus nullum civitatis nec senatus nec plebis concilium nec magi stratus'. But it is most unlikely that this is technical or refers in any way to Juristic Personality. Corpus nullum civitatis=no organised community. classical lawyers, is nowhere else expressed by any Latin word or phrase.

The other two alleged technical senses melt into each other, though logically distinct. For if some bodies were granted corporate character while others were not, it must have been because the State was more favourably disposed towards them; and if the State gave privileges to useful bodies, those privileges were likely¹ to involve recognition of those bodies as capable of rights and therefore as Legal Persons. But neither sense seems to be supported by any substantial evidence or to accord well with the language of the texts. For if corpus means a specially privileged or exceptionally corporate collegium we should expect to find such expressions as corpora et alia collegia; whereas in fact such expressions do not occur, and on the contrary there are two clear texts in the Digest that make collegium a subdivision of corpus., besides others that suggest the same thing? These two are D. io. 4. 7. 3 (Ulpian): 'idem etin collegiis ceterisque corporibus dicendum erit'; andD.47.22.3.1 (Marcian): 'collegium vel quodcumque tale corpus'. This seems good evidence against the view that some collegia are corpora and others not.

Every collegium, then, is a corpus. Is every corpus a collegium ? And if not, what other bodies have lawyers in mind when they talk of cetera corpora ?

Dogmatism here is dangerous, and it is important to realise that texts making corpus wider than collegium are very few, texts equating them very many. In the vast majority, the only bodies in question are the burial clubs and guilds of workpeople and traders, all called collegia, whose legal position will be discussed below? Such collegia were already numerous under the Antonines, and ¹ This is not inevitable; privileges such as exemption from taxes could be given to members of a body quite irrespective of whether it was corporate or not.

* Such as D. 12. 2. 34. 1; D. 47. 22. 3. 2; D. 40. 5. 36. 2; D. 48. 18. 1. 7. See Schnorr von Carolsfeld, pp. rsof.

³               Chapters iv and v.

were a very important element in the political economy of Byzantium; and it is clear that both inscriptions and emperors call them freely, almost indifferently, collegia and corpora.¹ They are not so prominent in the Digest, but there are some texts where corpus seems synonymous with collegium? and others where collegium vel corpus seems to mean ‘a college, whether it calls itself collegium or corpus'.3

More to our purpose are the texts which make corpus a wider term than collegium and those which enumerate some of the other bodies which in some respects resembled collegia. Perhaps the most suggestive is D. 48. 18. 1. 7 (Ulp. L. vin de officio proconsulisy. *servum municipum posse in caput avium torqueri saepissime rescriptum est, quia non sit illorum servus, sed rei publicae, idemque in ceteris servis corporum dicendum est; nec enim plurium servus videtur, sed corporis'. This Schnorr von Carolsfeld^{[19] [20] [21] [22]} thinks interpolated (from idemque on) for three reasons. He finds fault with the phrasing: 'in ceteris servis corporum' is an ‘eigenartige Wendung’, and the Nachsatz idemque...corporis repeats too much and is stylistically bad. Such an argument cannot be refuted, but is not likely to convince anyone who does not already believe the passage interpolated on other grounds. He points out, truly, that similar rules are given

30 elsewhere¹ for towns only; but that is natural if they were worked out first for towns and extended to other bodies. And he maintains that in the time of Ulpian the idea of a body as legally distinct from its parts was not yet suffi­ciently developed to be used in explaining such a rule— a contention which can only be dealt with after tracing the whole history of corporate bodies at Rome. But the real reason for the charge of interpolation seems to be the learned author’s belief that the word corpus never includes towns? This belief is paradoxical in view of Ulpian, Lib. sing, regularum, 22. $: ^lNec municipia nec municipes heredes institui possunt, quoniam incertum corpus est, et neque cernere universi neque pro herede gerere possunt, ut heredes fiant.'l It is certainly not supported by the text under discussion; whoever wrote the passage, Ulpian or Tri- bonian, evidently regarded a tow’n as a kind of corpus.

D. 2. 4. 10. 4 is less explicit. Ulpian is here represented as saying: ‘ Qyi manumittitur a corpore aliquo vel collegio vel civitate singulos in ius vocabit: nam non est illorum libertus, sed rei publicae honorem habere debet et si adversus rem publicam vel universitatem* velit experiri, veniam edicti petere debet, quamvis actorem eorum constitutum in ius sit vocaturus.' Schnorr von Carolsfeld thinks it very likely that a corpore... civitate is interpolated for a civibus·, ‘denn nur dann ist das “illorum” zu rechtfertigen*. This exaggerates the difficulty of supplying ‘members’ with singulos f and if singulos is admitted, illorum can present

¹ D. i. 8. 6. i; D. 2. 4. 10. 4, if interpolated as Schnorr von Carolsfeld maintains. See next paragraph.

⁴               Op. cit. p. 151: ‘corpus als Inhaltsbegrilf umfasse in den Digesten die öffentlichen Verbände nicht’.

³               Schnorr von Carolsfeld disposes of this text by saying (p. 154)“ Hier ist corpus in rein formalem Sinne gebraucht; die Stadt ist ein unbestimmter Organismus (incertum corpus), d. h. die Zahl der Bürger wechselt stets, ja es ist nicht einmal feststellbar, wie viele es gibt, denn stets ist eine Anzahl im Ausland’. See p. 26, supra and p. 34, infra.

⁴  For the possible interpolation of vel universitatem, see p. 38, infra.

⁵               The word corporatus does not occur, except in inscriptions, till the fourth century.

CORPUS

31 no difficulty. There is really no ground for suspecting the words ‘a corpore aliquo vel collegio vel civitate'; but it is not certain how they should be translated. They might mean either ‘by a corpus or a collegium or a civitas’ or ‘by a college (whether called corpus or collegium) or a civitas’ or ‘by a body such as a collegium or a civitas’. In other words corpus may be contrasted with collegium or identified with it or may embrace both collegium and civitas. The meaning of the passage as a whole is the same in each case, but that of corpus is different; and if we regard this text and D. 48. 18. 1. 7 either as both genuine or as both interpolated, it is most natural to regard corpus here as a general word covering both a collegium and a civitas. Moreover, if we accept this use as classical, it will affect our interpretation of D. 10. 4. 7. 3, where Ulpian, after speaking of municipes, adds: 'idem et in collegiis ceterisque corporibus dicendum erit’. If towns are corpora., then cetera corpora means ‘ bodies other than towns and colleges’; if not, it means ‘bodies other than colleges’. What, in either case, are these ‘ other bodies ’ ? An answer is suggested by five texts not yet considered in this connection.

D. 4. 2. 9. 1 (Ulp. L. xi ad ed.'): ‘etideo sive singularis sit persona, quae metum intulit, vel populus vel curia vel collegium vel corpus, huic edicto locus erit.'

D. 29.2.25. 1 (Ulp. Z. viii ad Sabi)·.¹ Servus municipum vel collegii vel decuriae heres institutus manumissus vel alienatus adibit hereditatem'

D. 37. 1. 3. 4 (Ulp. L. xxxix ad ed.): ‘A municipibus et societatibus et decuriis et corporibus bonorum possessio adgnosci potest.'

D. 46. 1. 22 (Florent. L. viii Inst.): ‘...hereditas per­sonae vice fungitur, sicuti municipium et decuria et societas'

D.3.4. I. pr. and § 1 (Gaius, L. m aded.prov.): ‘Neque societas neque collegium neque huiusmodi corpus passim omnibus habere conceditur...ut ecce vectigalium publicorum sociis per­missum est corpus habere vel aurifodinarum vel argentifodi-

narum et salinarum... § i. Qyibus autem permissum est corpus habere collegii societatis sive cuiusque alterius eorum nomine....’^I

These texts show that their authors were capable of associating with collegia (or corpora] the populus, the curia (once each), decuriae (three times) and societates (three times). Populus is a vague word under the Empire, when the Sovereign Assemblies had ceased to meet/ and the curia, or municipal senate, is not very like a collegium? But some decuriae are very like collegia indeed. These are the decuriae apparitorum or organisations of inferior civil servants, lictors, scribes and the like, which existed both at Rome and elsewhere and were normal collegia in all respects except for their connection with the State at a time when other collegia were wholly independent.⁴ It seems reasonable to suppose these are the decuriae referred to here, especially as we know from inscriptions that they owned and manumitted slaves.5

The societates quibus permissum est corpus habere will be discussed later.⁶ Here it is enough to point out that according to Gaius? they were able to own property and bring lawsuits; and that gives them a very striking re­semblance to collegia.

¹               For a full discussion of this text, see pp. 141 ff, infra.

¹ Schlossmann, p. 127, argues convincingly from D. 40. 9. 17. pr. that populus in D. 4. 2. 9. 1 means merely a mob. Cf. frag. Dosith. 7 (Girard, Textes, p. 507).           ³ But see p. 43, infra.

⁴               See Waltzing, vol. 1, pp. 54f; and for inscriptions, vol. iv,pp. 128—31. Decuria is also commonly used in inscriptions to mean one of the sections into which a collegium was often divided, Waltzing, vol. 1, pp. 358—61; vol. iv, pp. 291—304. Schnorr von Carolsfeld (p. 219) tentatively follows Heumann-Seckel in taking decuria to mean ‘town council’. This would be a natural meaning, since town councillors are regularly called decuriones\ but it is supported by no texts other than D. 29. 2. 25. 1 and D. 37. 1. 3. 4 (quoted above) and is not given in the Thesaurus Linguae Latinae. It seems however to occur in C. Th. 8. 5.46 (a.d. 385) and 14. 3.18 (a.d. 386) and may be a post-classical usage. Vat. Frag. 158 (Severus), ^lcollegae in decuria vel corpore', is ambiguous.

³ Waltzing, vol. 1, p. 55, n. 2.

⁶Pp. 143 ff, 159 ff., infra.                      7 D. 3. 4. 1. pr. and § 1.

From all this evidence it appears that corpus is never a technical term meaning ‘a corporate body’—even if it does once mean ‘corporate character’; but that classical Roman lawyers were quite aware that some bodies, groups, were recognised by law as able to own and sue, and in that respect as different from other bodies; and that conse­quently, when they talk of corpora, they tend to have these ‘corporate’ bodies primarily in mind. Sometimes corpora means only the guilds, collegia, which could own and sue, not because other bodies are not corpora, but because other bodies are not relevant. Sometimes decuriae and societates are relevant, and included. Corpus probably occurred with this extension in the edict, though the evidence of its being there at all is slender.¹ Municipia were already dealt with, and bodies not able to sue or be sued were irrelevant, so that if there was any such rubric as ‘ Qyod cuiuscumque corporis nomine agatur' or ‘ Quibus permissum est corpus habere—quod eorum nomine agatur', it must have referred to the collegia, and perhaps to decuriae and societates. Sometimes the extension is greater, and corpora stretches so as to cover bodies of men politically organised, towns,³ the State,3 the Senate,⁴ the plebs5 All are bodies, and what bodies are in contemplation can be learned only from the context, and seldom with precision, because the writer’s own thought is not precise. But since lawyers are mainly concerned with lawsuits, potential, if not actual, the bodies they are most likely to contemplate are those that can bring or defend actions; and when they want to talk about such bodies, the most convenient word is corpora.

Schnorr von Carolsfeld⁶ draws a different conclusion. He finds that corpus has three meanings:

‘i. Jede einigermassen feste, wenn auch nur soziolo- ¹ See Lenel, Das Edictum Perpetuum, 3rd ed., 1927, p. 100.

¹        See pp. 29-31, supra.

color=black face="Times New Roman">S   See p. 34, n. 7, infra.

⁴        Seneca, Epist. 102. 6.

⁵        D. 1. 2. 2. 25 and 26; Livy 1. 17. 2 and 4. 9. 4.

⁶         Op. cit. p. 175.

34 gisch organisierte Menschengruppe, abgesehen vom Staat und Gemeindeverband (Inhaltsbegriff).

‘ 2. Die Form einer soziologischen Organisation (Form­begriff, der auch die beim Staat und bei den öffentlichen Körperschaften sich findende Organisation mit um­fasst).¹

‘3. Technisch-eng: Verbände, insbesondere auch der Handwerker, später auch Zwangsverbände.’

The last sense is clear, and whether it can properly be called technical or not does not matter. But the distinction between the first two senses is difficult either to under­stand or to accept. The words ‘ Inhaltsbegriff’ and ‘ Form­begriff’ do no doubt convey some meaning to a German jurist, but it would take him a long time to explain them to Ulpian. And whatever they may mean, the first sense is curiously arbitrary. Even if we admit that where corpora include populus, legio^ grex² plebsj> the patricians,⁴ a group of slaves delimited by a legacy? or by their employ­ment in tax-collecting,⁶ the word is used in a ‘formal’ sense, it is very hard to believe that corpus (as Inhalts­begriff) could mean, as Schnorr von Carolsfeld seems to assert, a nation, a class, an army, any body of men organised ‘sociologically’ or forming a group of any kind, with the two exceptions of State? and Town. Common sense requires what we actually find in the texts: corpus can be used of any group, a guild, a town, or any other; and if it is commonly used by lawyers to mean ‘a group other

¹ ‘Dazu noch der Formbegriff corpus (hier vor allem der Unterbegriff corpus ex distantibus) im Sinne der Stoa, der aber auch für andere Gebilde als menschliche Verbände Anwendung findet.’ ² D. 41. 3. 30. pr., 2.

³  D. 1. 2. 2. 25 and 26; Livy, 4. 9. 4.

⁴  D. 1. 2. 2. 26.                                 5 D. 32. 38. 5.

6 D. 50. 16. 195. 3.

7       One text embarrasses him slightly (p. 163): ‘Als Inhaltsbegriff findet sich allerdings corpus in dem Sinne von Staatsverband bei Festus p. 158 s.v. Mamertini: in suum corpus, communionemque agrorum invitarunt (sc. Messanenses) eos (sc. Mamertinos). Es ist aber auch hier nicht der Staat als solcher, sondern die Eingliederung in den Organismus, von der die Rede ist.’

than a town’, that is only because they are following the Edict, which dealt first with towns and then with other ‘bodies’.

Another question discussed by Schnorr von Carolsfeld in connection with the word corpus is the extent to which the Romans—and the Greeks—conceived a body as distinct from its parts. Did the Stoics—did ordinary people in daily speech—regard corpora., groups of men, as Einheiten?^[XXIII] as Ganzheiten?³ as von der Summe der Teile unterschieden ?3 Thirty pages,⁴ packed with quota­tions, lead to the conclusion^ ‘dass man eine juristische Person (im Sinne der communis opinio der heutigen Rechtsdogmatik) in Rom noch nicht kannte, wohl aber eine besondere Art Gesamtpersbnlichkeit’. We must be grateful for these researches, which render valuable service in showing the linguistic and philosophical soil in which the lawyers’ usage was rooted; and we may thankfully accept the conclusion that the meaning which corpus seems to bear in legal texts is a meaning it can bear, in the light of philology and philosophy. Few Roman lawyers were great hands at either, and we need not suppose they could have answered, or would have understood, Schnorr von Carolsfeld’s questions.⁶ What matters is that, if the ques­tions and the possible answers had been explained to them, the answer he suggests would probably have seemed to them sensible.

UNIPERSITAS

None of the three words so far considered ever really means a ‘juristic person ’, in the technical sense of a ‘ right and duty bearing unit’. The nearest Latin equivalent to

36 this is ‘ Universitas' It does not quite correspond, since

it sometimes includes groups which are not ‘Persons’ and always excludes ideal Persons such as the fiscus (if it was a Person) and of course individuals; but it is nearer than persona, caput, or corpus.

Universitas originally meant a whole as contrasted with its parts, or, which is the same thing, a group as contrasted with its members; and this idea is never far away from any of its uses. Sometimes the contrast is strongly marked, sometimes it is quite latent. Where universitas is emphatic it may usually be translated ‘the thing as a whole’,² or ‘the group as a group’,3 often ‘the whole thing’⁴ or simply ‘the whole’.5

We are concerned here only with groups of men, not with aggregates of property. For though a hereditas is called a universitas, by Justinian if not by classical jurists, that is only because it is a number of things, including the res corporates hereditariae and later the debts due to and perhaps from it, regarded as a whole. We have seen that the Romans did not call it a persona·, nor did they

1       For an exhaustive but not altogether satisfactory study of this word see Schnorr von Carolsfeld, pp. 59—146. His very complete account is inter­esting and valuable, but mainly concerned to prove that universitas is not used by any jurist or lay writer before the Byzantine age, apart from Gaius 2. 11 and the suspected passages of the Digest, to mean ‘Körperschaft des öffentlichen und privaten Rechts’. It is not likely to have been so used, as a legal term of art, by any but lawyers, either in the second or the sixth or any other century; laymen do not often use words in technical legal senses. The lay use of universitas to mean a group of men or community does seem to have been rare in classical times, common in Byzantine; but this promising argument for interpolation shipwrecks on G. 2. 11, which cannot be ex­plained away. See p. 47, infra. Moreover, in the legal texts Schnorr von Carolsfeld seems over-subtle in his analysis. He finds it contradictory that universitas should mean different kinds of groups in different passages, sometimes including towns, colleges and even the State, sometimes only one or two of those classes. The truth seems to be that universitas was never really a technical term, and corresponded to our ‘group’ or ‘community’ rather than to ‘corporate body’ or ‘Legal Person’.

2       D. 29.1.18. pr.; D. 31.10; D. 34. 2. 8; D. 41.1. 7.11; D. 41. 4. 2. 6.

3G. 2. 11.                                      ⁴ D. 10. 2. 30; D. 50. 16. 239. 8.

⁵ D. 13. 4. 10; D. 46. 3. 73.

regard it as a Person. The use of universitas rerum (properly a composite whole, such as a flock of sheep¹) to mean a juristic Person consisting purely in funds earmarked for some purpose,² a Stiftung in the most technical sense, is a purely modern turn of phrase.                                              _a

A universitas^ then, is a group of people, with more or less emphasis on the fact that they are a group, and not so many separate individuals. By a natural development universitas comes to be used in particular by lawyers in speaking of a group which has been recognised by the law to be a group and either made capable or recognised to be capable as such of rights and duties. In other words, universitas sometimes means a collective juristic Person. This may be seen most clearly in the famous text of Ulpian which is the high-water mark of Roman thought on Legal Personality^ ‘si quid universitati debetur, singulis non debetur- nec quod debet universitas singuli debent'

It has been maintained by Albertario,⁴ Schnorr von CarolsfeldS and others that this use of universitas is post- classical, and that all the texts in the Digest which show such a use are interpolated. It is said that, except for three passages,⁶ where universitas stands for universus ■populus, universi cives, the classical jurists never use the term universitas without saying of what concrete universitas

¹               D. 7. i. 70. 3.

²               Girard, Manuel Elimentaire de Droit Romain, 8th ed., 1929, p. 256.

³               D. 3. 4. 7. 1; for its genuineness, see p. 45, infra.

⁴               Actio de Universitate e Actio Specialis in Rem, Rome, 1919. It had

before been suggested by H. Kruger, Z.S.S. vol. xxix, 1908, pp. 519—23, that classical lawyers used the word universitas to mean a town, whether a municipium, colonia, or peregrine civitas, and perhaps also to cover vici, pagi, praefecturae, fora, conciliabula, and the State itself, but not private corporations, Vereine. This view is more reasonable than Albertario’s, but involves unnecessary changes of text, and seems to rest on the common German assumption that the lawyers of ancient Rome, like those of modern Germany, had a ‘technische Sprache’ in which the same word always meant the same thing. This is not true of English legal terminology, and does not seem to be of either Roman or Byzantine.   5 See p. 36, n. 1, supra.

⁶               G. 2. n=D. 1. 8. 1. pr.; D. 1. 8. 2. pr. =Inst. 2. 1. pr.; and D. 1. 8. 6. 1 =Inst. 2. 1.6. See p. 47, infra. they are speaking—‘ senza dire di che concreta universitas si tratta’.¹ To prove this Albertario goes through what purports to be a complete list’ of passages in the Digest where universitas is used to express the juristic personality of a corporation, and gives reasons for thinking, or at least urges us to think, they have all been altered.3

Two questions have to be answered—what the word means in these passages, and whether it is interpolated. Albertario and Schnorr von Carolsfeld seem to find in the work of the compilers a technical sense foreign to classical usage. We would suggest that universitas was never a technical term and could always be used of any group, corporate or not, though when lawyers emphasise that they are talking of a group ‘as a group’ they normally mean ‘as a corporate body’. This implication is as clear in Gaius 2. ii as in any text of Justinian; conversely it was still possible in the fifth and sixth centuries to use universitas of an unincorporate group or community.⁴ Hence in going through Albertario’s texts we shall be prepared to find some interpolated and some not, and shall not, with Schnorr von Carolsfeld, regard the precise shade of meaning attributable to universitas as a useful criterion of a passage’s date.

D. 2. 4. 10. 4 (Ulp. L. v ad ed.'): ‘ ^ui manumittitur a corpora aliquo vel collegia vel civitate singulos in ius vocabit: nam non est illorum libertus. Sed rei publicae honorem habere debet et si adversus rem publicam vel universitatem velit experiri, veniam edicti petere debet, quamvis actorem eorum constitutum in ius sit vocaturus.'^

¹ Albertario, op. cit. p. 22.                        ² Op. cit. pp. 13-19.

3       Lenel, p. 100, n. 5, accepts Albertario’s conclusions as ‘sicher oder wahrscheinlich’.

4       Contrast C. 1. 23. 7. 2 with C. 2. 58. 2. 5, and see Schnorr von Carolsfeld, pp. 68ff. His contention that G. 2. 11 is interpolated (p. 91) carries no conviction.

5       On the first sentence of this text, see p. 30, supra. as if the words had been added by someone, glossator or compiler, who thought res publica not a wide enough term to cover collegium as well as civitas. It is clear from inscriptions¹ that collegia did themselves borrow from towns and the State the expression res publica, so that the deletion of vel universitatem leaves quite a satisfactory text. But we cannot suppose Ulpian’s voluminous works to have been free from awkward expressions, and it may have been an afterthought of his own that res publica was scarcely appropriate, in a legal treatise, to bodies which could only be called public in a very loose sense. The text may well be interpolated, but the evidence is not conclusive. We may translate: ‘But he must treat tne community with respect, and if he wishes to bring an action against the community, or group, he must ask exemption from the edictal rule, although he is not going to summon the community itself into court, but their appointed representative.’ The plural pronoun (eorum} can replace a collective noun as easily in Latin as in English.

D. 46. 8. 9 (Ulp. L. ix ad ed.}·. ‘Actor a tutore datus omnimodo cavet: actor civitatis nec ipse cavet, nec magister unruersitatis, nec curator bonis consensu creditorum datus.'

Albertario finds it strange that the actor of a corporation should be called magister·, but magister is the regular word for the president of a college,³ who took the chief part in all matters of administration and was the natural person to appoint as the college representative in a lawsuit. Cujas’ proposal, approved by Albertario, to omit universi­tatis and take magister to be the magister bonorum obsolete under Justinian, is extremely neat and may be right. But this change does not help with the most serious difficulty of the passage, which is the word ipse. The proposed omission leaves it completely in the air. As the text stands, we may construe ‘the actor of a town does not give security himself, nor does his analogue the president

¹                   E.g. C.I.L. vi. 9254; 1682; 9626; hi. 7485; xiv. 2299.

^{New Roman">2}               Waltzing, vol. 1, pp. 385—405. of a college’, though this is, it must be admitted, some­what strained. We cannot insert ‘ nec -procurator {cognitor) ab eo datus', the most natural antithesis, for an actor could not appoint a procurator? The most satisfactory, though speculative, explanation of ipse is to suppose that some such words as nec decuriones have dropped out after cavet. The actor of a town is no ordinary representative of an ordinary principal, and neither he nor his quasi­principals are compelled to give security. After decuriones, nec magister collegii would follow quite naturally, and it would be easily understood that the college committee were also exempt, like the decuriones. Ulpian may well have written either collegii or corporis. Universitatis is certainly less natural; we are compelled either to admit that universitas is used to mean a ‘ private law corporation’, i.e. a college, decuria,^{[24] [25]} or societas, as contrasted with towns; or to suppose an ellipse of 'alius', ‘other’, much easier in Latin than in English—‘the actor of a town does not give security himself, nor do the councillors; nor does the representative of any (other) group’. On this view, magister does come in awkwardly. Neither course is quite satisfactory, but both are possible; the case for holding universitatis to be interpolated is good, but not conclusive.

D. 3. 4. I. 3 (Gaius, L. in ad ed.prov.)·. 'Et si extraneus defendere velit universitatem, permittit proconsul, sicut in privatorum defensionibus observatur, quia eo modo melior condicio universitatis fit.'!

Here the only trouble is that the defendant body has been called eos and eorum in the last sentence but one, and Gaius could not ‘passare improvvisamente dal concreto (membri della corporazione) all’ astratto (personalità giuri­dica della corporazione)’.^[26] But the Romans did pass with the greatest ease from concrete to abstract and vice versa. In D. 2. 4. 10. 4, on Albertario’s own interpretation, eorum refers to rei -publicae and should logically be eius, an exact parallel to this passage. And municipes is much commoner than municipium as an equivalent of civitas ;* especially striking are D. 37. 1. 3. 4: ^la municipibus et societatibus et decuriis et corporibus bonorum possessio adgnosci potest', and D. 34. 5. 2, where Papinian says 'civibus civitatis legatum vel fideicommissum datum civitati relictum videtur'. This terminological inexactitude is not peculiar to the Romans: the corporation commonly called Trinity College, Cambridge, is officially ‘The Master, Fellows, and Scholars’; the town of Cambridge is ‘The Mayor, Bailiffs, and Burgesses’? The text states clearly a reason­able rule. Universitatis fits perfectly into its context, if it is allowed to mean ‘a group’ of the kind under discussion, i.e. a corporate societas or collegium. Only a perverse belief that universitas must always bear a uniform meaning, always including or always excluding towns, could lead to suspicion of so straightforward a passage.

D. 3. 4. 2 (Ulp. L. viii ad ed.): ‘ Si municipes {-pet F) vel aliqua universitas ad agendum det actorem, non erit dicendum quasi a pluribus datum sic haberi: hic enim pro re publica vel universitate intervenit, non pro singulis'

Albertario objects to the concrete municipes beside universitas, and suggests that the original reading was municipium, but was made hard to read by the interpolation following. This is ingenious, but his objection cannot be taken seriously in view of the passages cited in the last note. The singular verb may perhaps favour Albertario’s reading: but det {dent) for det is surely a small enough correction to be allowed even in a book with such a peculiar textual history as the Digest; if not, convenire et conveniri, cum ius omnium in unum recciderit et stet nomen universitatis.'

Albertario begins by objecting to 'In decurionibus vel aliis universitatibus' as nonsense, ‘cosa senza senso’, on the ground that the decuriones do not form a universitas. This seems plausible, but is not quite self-evident. The decuriones do not seem to have owned property as a group, but that of the town is sometimes ascribed to them¹ and confusion between the corporate Personality of a town and that of its governing body, its ‘corporation’, is the most natural thing in the world. Moreover, the curia has some important attributes of a juristic Person; it retains its identity, as Ulpian says, through all changes of members, and it takes decisions by the vote of a majority. The decuriones do not appear as such in litigation, but they have enough of a collective will to appoint the actor who represents the town. They may really be said to form a Person, though it is a Person of public law rather than private. And we cannot tie Ulpian down to any modern definition of universitas restricting it to private law corpora­tions. A town council is a body of men acting, and regarded, as a whole which is distinct from its members; and that is enough to justify his use of the word: we may translate ‘ In the case of town councils and other groups.’’

If on the other hand we admit Albertario’s contention, and excise vel aliis universitatibus, the remains of the sentence are unsatisfactory. ‘It does not matter whether all the town councillors remain the same, or some remain, or all change’ is a statement which has very little meaning or interest if the decuriones are not regarded as a universitas. For this is one of the fundamental characteristics of a true corporation, and it is incredible that in a text where at ^{[29] [30]} least the towns are being discussed as corporate bodies Ulpian should, as Albertario would have us believe, have confined this vital rule to bodies which he did not consider corporate at all. We may also notice that this section follows one from which no change of words can banish the idea of corporate character; and that while we should naturally expect a further rule about universitates, nothing could be less appropriate than a reference to a set of unincorporate officials.

The last sentence (Sed si...stet nomen universitatis') Albertario would remove altogether. He says that sed has no sense, and this seems to be true, though it may conceivably be used for sed et or sed etiam, meaning ‘yes, and what is more’, ‘nay more’, ‘moreover’.¹ But the compilers in their haste were far more likely to miscopy a conjunction than to add a sentence. 'Magis admittitur' is objected to by H. Krüger as ‘sprachlich verdächtig’, but such arguments are now rightly suspected. Both H. Krüger and Gradenwitz find ‘incongruente’ the phrase 'ius omnium' after universitatis, but ‘members’, for which there is no Latin equivalent, can very easily be understood. ‘Eorum', in 2. 4. io. 4, which Albertario does not mind, is far more incongruous.² This is all the evidence he adduces against the sentence or against the use of Universi­tas-, and it does not seem very impressive. SiberS points out that if the sentence is not to be struck out altogether, it must refer to collegia, not to towns or town councils, which are not likely to be reduced to a single member. Consequently Universitas in this section cannot be simply omitted (Gradenwitz and P. Krüger), or replaced by civitas (Albertario and Arangio-Ruiz). Siber would write collegium for universitas·, but the more general word gives more coherence to the text.

We now come to the central and important § 1: ¹ si quid

¹ Cf. sed et legato, h. 1. pr.                        ² See p. 38, supra.

³               H. Siber, Römisches Recht in Grundzügen für die Vorlesung, n, Römisches Privatrecht, Berlin, 1928, p. 52, n. 12. universitati debetur, singulis non debetur: nec quod debet universitas singuli debent'. This certainly does not join on very closely or naturally to the preceding sentence; if we preserve the reading, we may suppose a paragraph of Ulpian to end at ‘actionem in municipes'—it is quite likely that something is omitted—and another to begin at ‘si quid universitati'. This paragraph then runs on quite naturally to the end of the fragment. Albertario does not suggest such a break, but points out that we have four other extracts¹ from the same context (Ulp. L. x ad edl) dealing with municipes, and none mentioning any other universitas. He therefore proposes either to read: ‘si quid civitati debetur, singulis civibus non debetur: nec quod debet civitas cives singuli debent'; or to take universitas as equiva­lent to universi cives.

The second plan is quite satisfactory. If universitas can mean a group, it can certainly mean the group in question. But it must be clearly recognised that universi cives means ‘the citizens as a group’, not ‘all the individual citizens’. His other proposal also gives good sense, and it seems clear, though his other extracts from Ulp. L. x ad ed. are very short, that the context was mainly concerned with civitates. This, however, proves little. It is quite compatible with Lenel’s view^{[31] [32] [33]} that the Edict contained, besides the rubrics Qyibus municipum nomine agere liceat and Quod adversus municipes agatur, a rubric on actions brought by and against corporations other than towns, probably calling them corpora. The first of these rubrics he places at the end of Ulpian’s ninth book on the Edict, the other two being treated in the tenth, and all three in the ninth book of Paul ad edictunfi and the third of Gaius ad edictum pro­vinciate. He points out^[34] that the rules for towns and other corporations are given separately though consecutively in the texts.¹ Thus Paul gives for each in turn the rule that a filiusfamilias can be an actor? Gaius tells us that uni­versitates could,3 Javolenus that towns could not,^[35] be defended by an extraneus; and we find in h. t. 2 the apparent contrast municipes vel aliqua universitas, pro re publica vel universitate. It is possible that universitas came to be used in practice to mean a group or corporate body other than a town,^[36] but it has been urged above⁶ that this interpretation is not necessary in h. t. 2. Lenel’s evidence certainly suggests that so neat and clear an arrangement is not due to Tribonian, even if it does not conclusively prove that towns and colleges were not dealt with consecu­tively under one rubric. It is possible, though scarcely probable, that the rules for universitates other than towns are due neither to the Edict nor to Tribonian, but to the classical commentators. We have no text of Julian to show what rules there were in his time, and we cannot really tell what is Edict and what is comment. But we may note that Gaius, the earliest of our three authorities, has the simplest, most fundamental, definition of corporate cha­racter, 7 while Ulpian, the latest, gives us with admirable clearness of thought and expression some of the less obvious, though essential, principles.⁸ This development makes it unlikely that these texts owe anything to the commentators of Berytus. In any case, the compilers can hardly have interpolated the whole of fr. i, and there is no good reason for thinking that they either inserted bodily the other rules applying to colleges or skilfully adapted rules that originally only covered towns. It does not indeed follow that because the matter is classical the wording must be: most of the compilers’ changes were purely verbal, and in all these passages we could either omit the word universitas or substitute collegium or corpus without serious damage to the sense. But the probabilities seem against it.

and it seems precisely to signify the corporate character which Albertario denies. Gaius clearly has corporate character in mind, as he has in D. 50. 16. 16: ¹¹ civitates enimprivatorumlocohabentur''theobscureD.3.4. i,3 where he is discussing the corporate character of some bodies.’4 Again, in D. 1. 8. 2. pr. there is an antithesis which Albertario does not explain. Marcian says: ‘ Quae­dam naturali iure communia sunt omnium, quaedam uni­versitatis, quaedam nullius, pleraque singulorum.' Here again universitas must mean ‘a body’ rather than ‘everybody’;5 for if it means ‘all the people concerned’, universi cives or universus populus according to the context, there is no

48 difference between universitatis and omnium. Only in D. 1. 8. 6. 1. is Albertario’s explanation satisfactory and universitatis equivalent to civium universorum.

It appears from the foregoing pages that Albertario has not conclusively proved any of these passages Byzan­tine, though he has shown cause for suspecting some of them. It seems likely, though not certain, that the classical lawyers, though probably not the edict, did at least some­times, though not often, use the word universitas to denote a corporate body, a group recognised by the law to be a group, and not merely so many individuals. Justinian uses the word oftener, but not much oftener, and in very much the same sense. The compilers are more familiar than the classical lawyers with the idea that towns and collegia are in some respects alike, and more ready to lay down for ‘all groups’ rules originally stated separately for groups of different kinds. But it is a perverse anachro­nism to suppose that Tribonian’s use of the word implies any consciousness of the innumerable subtleties conveyed to modern German minds by the words ‘juristische Person’.

CONCL US IONS

The foregoing pages may throw some light on the question what words were available to classical and Byzantine lawyers when they wanted to write, or speak, or think, about Legal Persons. We have seen that caput was used from early times to denote the legal position lost on enslavement; but the use remained negative, and we never hear even of a slave’s acquiring caput on manumission, much less of a town or a college having caput. Persona comes in Byzantine times to be an attribute which litigants must possess, and even under Domitian a town could be called a persona. But it is clear and certain that neither word was actively present to the minds of lawyers when they were working out the legal position of towns and

colleges, the first Legal Persons other than individual men to be clearly recognised as right-and-duty-bearing units, as actual or potential litigants. Most of their mental operations were performed on, or with, the clear-cut words civitas, colonia, municipium, collegium. It might be hard to say what a municipium could or could not do, but whether a given community was or was not a municipium was as straightforward a question of positive law as the status of an English town as a Borough or an Urban District. Two words, however, allowed some elasticity, namely corpus and universitas. Corpus is certainly the more familiar term to classical lawyers, and probably appeared in the Edict. It means 'collegium or other entity of the kind’; and the question what bodies are of the same kind as collegia brings fairly into issue the nature of corporate character, and can be translated into classical Latin. What­ever view is taken of D. 3. 4. 1, it cannot be doubted that Gaius or Julian, if asked ‘ Quibus corporibus licet ad exemplum rei publicae habere res communes et actorem per quern quod communiter agi fierique oporteat agatur fiat?' would have understood the question and been prepared to discuss it. Roman lawyers, like those of England, were in no hurry to generalise their statements, and they seem to have built up a good deal of law about towns and colleges without having any word to mean either ius rerum communium et actoris habendorum or municipium vel corpus cui licet res communes et actorem habere. England has a great deal of law about corporations, yet neither phrase can be trans­lated into lawyers’ English. ‘Corporation’ and ‘corporate capacity’ are too narrow. ‘Legal Person’ is a foreign- sounding phrase, ‘juristic Personality’ still more so. The apparatus of Trusts¹ often saves us from naming entities which to a German eye must appear as Persons, and we are content on occasion to help out the inelastic Corpora­tion with so absurd an expression as Unincorporate Body; but when we are brought face to face with a right-and-duty- ¹ See chapter ix, infra.

50 bearing unit which is clearly not a Body at all, such as a Hindu idol, we take refuge in sham Roman law and call it a -persona.

No Latin word or phrase is synonymous with ‘juris­tische Person nor did the Romans ever mark off in their minds the area covered thereby. But they certainly did make the cruder generalisation implied by the English word ‘corporation’. Certainly Tribonian, and almost cer­tainly Julian, knew very well that towns and guilds were somewhat alike, and that their essential points of resem­blance were the right to own and the right to sue.