. 42
( 55 .)


also be the manuscript copied by or for Pierre Hevin, as one of the two versions of the Assize he
kept. This copy is headed, `Ex Cod. D. Petavi/ Haec est assisia terrarum, quam fecit
GAUFREDUS Comes ®lius Regis Anglorum' (AD Ille-et-Vilaine, `Fonds Hevin', 1F 5).

Appendix 1
complained that he had not been able to see this manuscript because the Vatican
did not answer his letters. At that time, he hoped it was the original Dinan
manuscript, and therefore that one original baronial manuscript was extant. In his

edition of the `Tres ancienne coutume' published a few years later, the manuscript
is described as a thirteenth-century copy, so presumably Planiol had had more
success with the Vatican but his hopes were dashed.13 A French translation of the

Dinan text was also incorporated into manuscripts of the `Tres Ancienne Coutume
de Bretagne' in the mid- to late ®fteenth century.14

4. Leon

The earliest known manuscript of the Leon text is incorporated in a manuscript of
the `Tres Ancienne Coutume' dated 1437.15 The text is corrupt, and it is probably

the text of the Assize which has suffered the most from later scribal errors (e.g.,
`voluerit' for `noluerit' and vice versa).

5. Porhoet
The only copy of the Porhoet text is that preserved in the fourteenth-century
cartulary of the abbey of Saint-Melaine de Rennes.16 Identi®cation of this text
with Eudo de Porhoet is not entirely certain, since the addressee's name appears
only as `e. comiti'. No other `E . . . comes' is known in Brittany in this period,
though, and Eudo de Porhoet is named as a witness in some of the texts, styled
`comes Eudo'.

6. Rohan
The earliest manuscript of the Rohan text represents the earliest extant copy of any
text of the Assize. As noted above, the Rohan text was entered in the second
register of documents made for Philip Augustus between 1212 and April 1213.17
Since this occurred less than thirty years after the Assize was made, it is quite
possible that the royal clerks had access to the Rohan `original'. Another
thirteenth-century copy of the Rohan text was preserved in the archives of the
abbey of La Vieuville.18 This could equally have been the source of the royal copy,
but the relationship between the two thirteenth-century copies and the Rohan
`original' cannot now be determined.

Planiol `Assise' at p. 126, TAC, p. 321.

Bibl. mun. de Rennes ms 15964 (published in Planiol `Assise', pp. 123±4); J. Brejon de

‚ ‚
Lavergnee (ed.), `Une version francaise inedite de l'Assise au comte Geoffroi', in Recueil de
‚ ‚ ‚‚
memoires et travaux publie par la Societe d'histoire du droit et des institutions des anciens pays de droit
‚ ‚
ecrit, fasc. VII, Melanges Pierre Tisset, Montpellier, 1970, pp. 65±75 at 74±5. For other references
see TAC, p. 320. Another French translation is at BM additional ms 8876, folios 159±60.
TAC, p. 319.

`Cart. St-Melaine', fol. 183; BN ms fr. 22325, p. 73.

AN ms JJ8, fol. lxiii v, no. 297; J.W. Baldwin (ed.), Les Registres de Philippe Auguste, I, Texte,

Paris, 1992, pp. 555±6.

BN ms fr. 22325, pp. 571, 607. The copy at p. 607 (which bears the marginal note, `ecriture du

13 siecle') may have been a draft for the copy on p. 571.

The `Assize of Count Geoffrey'
The Rohan text has been widely copied throughout the centuries since 1185
and not only within Brittany, but also in Normandy and in Paris, and has been
published several times.19 Brejon de Lavergnee has published a Parisian copy of the

Rohan version, in French and so corrupt that its only interest is as a source for the
reception of medieval legal texts in the later middle ages, rather than as a source for
the Assize itself.20 For this, the extant manuscript of 1212 is of primary importance.
It is unfortunate, then, that in copying the Rohan text, presumably in the interests
of brevity, the royal clerk omitted parts of the text which were not material to the
substantive provisions, that is, the clause `Notum sit . . . quod' and the witness-list.

The ®rst clause occurs in all the texts except Rohan and Vitre, and is consistent
with the diplomatic of Duke Geoffrey's acta. It remains possible, however, that this
clause was omitted from the Rohan `original'.

7. Vitre

The Vitre text is the best source for the Assize, since there is evidence that the
original manuscript, with the seals of Duke Geoffrey and Duchess Constance
attached, survived into the sixteenth century, when it was used for two printed
editions, in 1536 and 1552 respectively.21

8. Anonymous
An `anonymous' text of the Assize appears in a late ®fteenth-century manuscript of
the `Tres Ancienne Coutume de Bretagne'.22 This text is `anonymous' because the

®nal clauses of the Assize, in which the name of the addressee appears in all the
other versions, have been omitted. Comparing this text with the others, it most

closely resembles the Vitre text, although it cannot be positively identi®ed without
the ®nal clauses. Curiously, a French translation of the Dinan text appears at folios
159±60 of the same manuscript, but the `anonymous' Latin text is not of the Dinan
text. The `anonymous' text may represent an eighth text, addressed to an
unidenti®ed eighth baron. Certainly, there is no reason why the seven barons
known to have acquired a copy of the Assize should be the ®nal number.

The texts of Vitre, Dinan and Rohan are the most useful, but not because these
are somehow `closer' to the supposed original `of®cial' text of the Assize and
therefore a better, more accurate record of it. Rather, the extant copies of these

three are, for the reasons given above, closer to the original manuscripts of Vitre,
Dinan and Rohan and less likely to have suffered later scribal error. The other four
(or ®ve) versions, in contrast, are ®rst recorded in later medieval manuscripts, in
contexts in which the provenance of the text is indeterminate and there is ample
scope for scribal error. The variations between the three most reliable texts,

AN ms J240, nos. 30±31; A. Teulet (ed.), Layettes du Tresor des Chartes, i, Paris, 1863, p. 144, no.

337. The Rohan text was published in M. Brussel (ed.), Usage general des Fiefs, Paris, 1750, book
iii, ch. xiii, p. 883, after a manuscript incorporated in the `Terrier Cartulaire de Normandie',
which was destroyed in 1737.
‚ ‚
Brejon de Lavergnee (ed.), `Une version francaise inedite', pp. 73±4.

Planiol, `Assise', 127, 120±2.

BM additional ms 8876, fol. 157.

Appendix 1
however, demonstrate that not all the variations between different texts of the Assize
can be attributed to error or deliberate glossing by later copyists. They prove that
variations existed between the original baronial copies of the Assize, but that the
variations were merely of word-order and did not affect the substantive provisions.
In view of my argument, it is perhaps misleading to seek to establish a single text
of the Assize. The ideal edition of the Assize would include a critical edition of
each of the seven (or eight) known texts. Each of these is derived from an original
`baronial' manuscript, made in or soon after 1185, and possibly sealed by Duke
Geoffrey. There does not appear to have ever been a single `of®cial' ducal text of
the Assize, of which these can be regarded as mere copies. If there ever was a ducal
text, it may have been no more than a draft prepared by ducal clerks and used as the
basis for the baronial texts, each of which became a legitimate text when
authenticated by attachment of the ducal seal.
Space does not however permit the editing of eight texts, and published editions
‚ Ã
exist (albeit not all to modern standards) of the Vitre, Dinan, Rohan, Chateaugiron

and Leon versions. It is, however, useful to have a text of the Assize which is
`complete' in that it has all the clauses common to the majority of the different
texts, and `correct' in that it omits the most obvious later glosses and errors. My
edition, attached to this Appendix, was prepared in this spirit.

the substantive provisions of the assize
To state my argument very brie¯y at the outset, the Assize does not represent the
introduction into Brittany of new principles of succession from England, Nor-
mandy, or elsewhere in the Angevin empire. Rather, it is a statement of the
customary law of baronial succession in Brittany. This is apparent from an
examination of the provisions of the Assize in their historical context, especially if
one looks beyond the initial statement of the principle of primogeniture and
considers the rest of the Assize, the clauses I have termed the `subsidiary provisions'.
As a preliminary matter, it is necessary to determine to whom, or to which lands, the
Assize applied. The Assize expressly states that it concerns baronies and knights' fees (`in
baroniis et feodis militum'). Neither expression for land-tenure is generally used in
other contemporary documents in Brittany. Probably these were Anglo-Norman
imports, used here as a convenient way of describing certain types of landholding. The
expression `feuda militum' does occur in the Inquest of Dol (1181), but this was
produced under Henry II's seneschal of Rennes, an Englishman.23
The words, `baro', `miles' and `feodum' are commonly used in Breton documents
of the eleventh and twelfth centuries, but with a wide range of meanings.24 In the
eleventh century, at least, a great magnate may be styled miles or baro, and a miles
may also possess allodial land. Equally, the charters of twelfth-century magnates like
Alan de Rohan and Ralph de Fougeres may refer to men who held what we would
call knights' fees as their barones. By the time of the Assize, the late twelfth century,

Enquete, p. 39.

See A. de la Borderie, Histoire de Bretagne, iii, Rennes, 1899, p. 283; N.-Y. Tonnerre, Naissance

‚ ‚
de la Bretagne: Geographie historique et structures sociales de la Bretagne meridionale (Nantais et
Vannetais) de la ®n du VIIIe a la ®n du XIIe siecle, Angers, 1994, pp. 365±406.

The `Assize of Count Geoffrey'
more social strati®cation had occurred, with the aristocracy divided into two: an
upper echelon of great magnates, the barones (whence barons), who possessed
extensive estates and multiple forti®cations in virtual autonomy, and the milites
(knights), who possessed their lands in close dependency upon either a baron or the
The Assize recites that it was made with the counsel of barons, and all those to
whom copies of Assize are addressed are barons. What, if any, was the role of the
knights, or, to be precise, the holders of knights' fees? Anyone holding a knight's
fee must have held it of some lord, either the duke (that is, as a tenant of ducal
domain) or a baron. The duke, and each baron who was a party to the Assize,
warranted that he would regulate the successions of his knightly tenants according
to the provisions of the Assize. It was, therefore, unnecessary for such tenants to
consent to the Assize or receive a written copy of its terms.

The custom of primogeniture in Brittany
The Assize is usually described summarily as introducing to Brittany the principle
of succession by primogeniture. The ®rst substantive provision of the text, `ulterius
non ®erent divisiones sed major natu integre obtinet dominatum', certainly ordains
primogeniture, but it does not necessarily introduce the principle. In fact, there can
be no doubt that succession according to the principle of primogeniture was
already the custom of baronial families. Even Arthur de la Borderie, who wished to
interpret the Assize as a tyrannical Angevin imposition, was obliged to admit this.
He thought that the Bretons adopted primogeniture when, after the Viking
Á Á ‚ ‚
invasions, `ils imiterent et importerent chez eux les institutions de la feodalite
francaise', but argued that the custom was contrary to ancient Breton law and that
the numerous and substantial divisions of baronies during the twelfth century
manifested an inherent distaste for the principle.25 Succession by primogeniture has
been demonstrated for the lords of Combour, Fougeres, Rays and Vitre.26 In fact,
Á ‚
any baronial family one chooses to examine will demonstrate succession by
primogeniture from its earliest appearance in the eleventh century.
Nevertheless, the division of baronial estates between sons or brothers did occur
before 1185, and such divisions are cited in support of the argument that
primogeniture was an innovation imposed by the Assize.27 In fact, these cases
prove quite the opposite. It is necessary to distinguish between a principle of
partible inheritance, which dictates that the patrimonial estates should be divided
between the deceased's sons (whether in equal shares or otherwise), and the

A. de la Borderie, Histoire de Bretagne, iii, p. 282.

‚ Á
H. Guillotel, `La devolution de la seigneurie de Dol-Combour aux XIe et XIIe siecles.

Contribution a l'etude des successions seigneuriales en Bretagne avant l'Assise au comte

Geoffroy', RHD, 4th series, liii (1975), 190; J.F. Auberge (ed.), Le Cartulaire de la seigneurie de
Fougeres, connu sous le nom de Cartulaire d'Alencon, Rennes, 1913, p. 54; R. Blanchard (ed.),
Cartulaire des sires de Rays (1160±1449), Archives historiques de Poitou, xxvii, Poitiers, 1898,
`Introduction', pp. l±lxxiii; M. Brand'honneur, `La lignage, point de cristillisation d'une
‚ ‚ ‚ Á
nouvelle cohesion sociale. Les Goranton-Herve de Vitre aux XIe, XIIe et XIIIe siecles',


. 42
( 55 .)