¹ Walthew, C.V., ‘Possible standard units of measurement in Roman military planning,’ Britannia xii (1981), 13–36.Google Scholar

² Blagg, T.F.C., ‘Roman civil and military architecture in the province of Britain,’ World Arch, xii (1980), 27–42.CrossRefGoogle Scholar

³ The other ancient source which should perhaps be mentioned is pseudo-Hyginus, De munitionibus castmrum. The author of this treatise is, however, concerned with the rational deployment of prefabricated entities (i.e. tents), and his preoccupations cannot be expected to be those of the architect who was not primarily concerned with how many people could be fitted into a given space.

⁴ Vitruvius, De architectura I.praef. 2; Liv.I.

⁵ G. Webster, The Roman Imperial Army (3rd edn, 1985), 158.

⁶ Vitruvius VII.praef. 10-18.

⁷ Thus Faventinus (fl. c. A.D. 300); K. Plommer, Vitruvius and Later Roman Building Manuals (1973), introduction passim. A distinction should, however, be made between the principles upon which the design process was based, and the details through which the individual architect expressed himself, see Blagg, T.F.C., ‘Reconstruction of Roman decorated architecture,’ in Drury, P.J. (ed.), Structural Reconstruction, BAR 110 (1982). 131–52.Google Scholar

⁸ Walthew, op. cit. (note 1), 18.

⁹ MacMullen, R., ‘Roman imperial building in the provinces,’ Harvard Stud. Class. Phil, lxiv (1959), 215.Google Scholar

¹⁰ e.g. ILS 2034.

¹¹ Millar, F., ‘Emperors, frontiers and foreign relations 31 B.C. to A.D. 378,’ Britannia xiii (1982), 5; idem, The Emperor in the Roman World (1977), 315f. See also J.B. Campbell, The Emperor and the Roman Army (1984), esp. 104-5 and 279 for the sort of situations in which an emperor might or might not take an active part in the minutiae of army life.Google Scholar

¹² See for example Hadrian's Wall, where it appears that the original plans were drawn up away from the site and had to be modified during construction (D. Breeze and B. Dobson, Hadrian's Wall (2nd edn, 1978), 38). We are in fact informed that Hadrian established a corps of architects and building specialists (Ep. de Caes. 14.15), but though they seem to have been organised on a quasi-military basis there is no evidence that they had military duties: indeed the building programme in Rome and the surrounding area would have been enough to keep a small army of specialists employed.

¹³ R. Wilson, Roman Forts (1980), 45.

¹⁴ Procopius, On the Buildings IV.61.12-13. For what is known of his career, see W. D. Macdonald, The Architecture of the Roman Empire I (2nd edn, 1982), 134; and F. Leppar and S.S. Frere, Trajan's Column: A New Edition of the Cichorius Plates (1988), 187-93.

¹⁵ It has been suggested by Zienkiewicz that the Fortress Baths at Caerleon might have been designed under such circumstances, as they are much closer to the tradition of the ‘Imperial’ baths than they are to normal legionary bath buildings in the north-west provinces (J.D. Zienkiewicz, The Legionary Fortress Baths at Caerleon (1986), 119 n. 25).

¹⁶ Cicero, ad Q.fr. II.3 III.I; ad Att. II.3, XIV.9; ad fam. VII. 14; Pliny, Ep. IX.39.

¹⁷ Pliny, Ep. X. 17b, 18, 37-42, 61, 62.

¹⁸ Pliny, Ep. X.40.3: ‘There is no province which does not have both experienced and gifted men’.

¹⁹ G. R. Watson, The Roman Soldier (1969), 144.

²⁰ Digest L.6.7.

²¹ RIB 1542, 2091, 2096.

²² Unlike those from the other provinces and Italy, recorded below.

²³ Architects did not normally sign buildings (Enciclopedeia dell'arte antica 1, 574). The Bassus or Bassianus whose name appears on the building inscription from Beauport Park as curam agens would have been the foreman of the ironworks (and a civilian) rather than the architect (contra G. de la Bedoyère, The Buildings of Roman Britain (1991), 15; see Brodribb, G. and Cleere, H., ‘The Classis Britannica bath-house at Beauport Park, East Sussex,’ Britannia xix (1988), 241, 261).Google Scholar

²⁴ CIL XIII.7945. For the expansion see Année Epigraphique 1942-3, no. 93. For evidence from the Praetorian Guard for apprenticeships, see M. Durry, Les cohortes pretoriennes (1938), 115-16.

²⁵ Miller, S.N., ‘Notes on an inscription from Birrens (Dumfriesshire),’ JRS xxvii (1937), 208–9.Google Scholar

²⁶ CIL X.630 and 1757, one of whom is described as veteranus.

²⁷ CIL VI.3182.

²⁸ CIL X.3392-3.

²⁹ CIL VIII.2850; CIL III.6178, Eph. Ep. IV.138 (IV.22).

³⁰ CIL XIII.8082.

³¹ C Vedennius C f / Qui(rina) Moderatus Antio / milit(auit) in leg(ione) XVI Gal(lica) a(nnis) X / translat(us) in coh(ortis) IX pr(aetoriae) / in qua milit(auit) ann(is) VIII/ missus honesta mission(e) / revoc(atus) ab imp (eratore) fact(us) evocfatus) Aug(usti) / architect(us) armament(arii) imp(eratoris) / evoc(atus) ann(is) XXIII / donis militarib(us) donat(ur) / bis ab divo Vesp(asiano) et / imp(eratore) Domitiano Aug(usto) Germ(anico) (ILS 2034=C/L VI.2725).

³² T Flavio T f Pup(inia) Rufo / militi coh(ortis) XII urb(anae) / et coh(ortis) IIII pr(aetoriae) / ordinato architec(to) / tessario in (centuria) / b(eneficiaro) praefector(um) praetor(io) / cornicular(io) praeflecti) anno(nae) / (centurioni) leg(ionis) XIIII Gem(inae) et XI Cl(audiae) et / II Aug(ustae) et VII Gem(inae) / Ulpia pientissima soror et/Aelia Secundina heredes/ex testamentum faciend(um) curav(erunt) (CIL XI.20).

³³ Pitts, L.F. and Joseph, J.K. St, Inchtuthil. The Roman Legionary Fortress, Britannia Mono. 6 (1985), 155.Google Scholar

³⁴ Davison, D.P., The Barracks of the Roman Army from the First to Third Centuries A.D., BAR S472 (1989).Google Scholar

³⁵ Breeze and Dobson, op. cit. (note 12), 56-64.

³⁶ This might be a long time. Vedennius Modestus (note 31) served for 41 years altogether.

³⁷ MacMullen, op. cit. (note 9), 215; idem, Soldier and Civilian in the Later Roman Empire (1963), 35.

³⁸ Pliny, Ep. X.40: ‘They have usually come to us from Greece’.

³⁹ E.W. Marsden, Greek and Roman Artillery. I. Historical Development (1969), 188, 192.

⁴⁰ A notable exception is Crummy, P., ‘The origins of some major Romano-British towns,’ Britannia xiii (1982), 130Google Scholar. For a detailed discussion of this problem, see Coulton, J. J., ‘Towards understanding Greek temple design,’ Ann. Brit. School at Athens lxi (1974), 59–99 (hereafter cited as Greek temple design). My debt to this article in this and the following section is obvious.Google Scholar

⁴¹ A further modifying factor, the possibility of ground movements which may have taken place between the erection of the building and its eventual excavation, should not be forgotten when the building is assessed, although this factor is not included here as it lies beyond the scope of this paper.

⁴² e.g. Vitruvius VIII.v.I; IX.vi.2.

⁴³ Vitruvius IX.viii.5.

⁴⁴ Vitruvius III.i.3.

⁴⁵ ibid.

⁴⁶ Grenier, A., Manuel d'archéologie gallo-romaine III (1958), 36–8; O.A.W. Dilke, The Roman Land Surveyors (1970), 73.Google Scholar

⁴⁷ Thanks are due to the National Museum of Wales for the opportunity to examine this piece.

⁴⁸ See below p. 151.

⁴⁹ ‘Asseres abiegi ad lineam aut regulam aequaliter dirigantur’ (Faventinus 21).

⁵⁰ Reproduced in MacDonald, op. cit. (note 14), 139, pl. 127.

⁵¹ e.g. Walthew, op. cit. (note 1), 22.

⁵² Suetonius, Div. Aug. 24.2.

⁵³ Dilke, op. cit. (note 46), 67, 73, quoting Nowotny, Römische Forschungen in Österreich (1925). These are published as being from a decempeda, but there is nothing to indicate that they could not have come from a large regula.

⁵⁴ The bronze tips from Ems give rise to the intriguing speculation that some equipment may have been designed to provide two different systems of measurements, like our own dual metric/imperial equipment.

⁵⁵ e.g. Walthew, op. cit. (note 1), 18.

⁵⁶ Vitruvius VII.iii.5: ‘longitudines ad regulam et ad lineam, altitudines ad perpendiculam, anguli ad normam respondentes exigantur’.

⁵⁷ Faventinus 21.

⁵⁸ Vitruvius IV.iv.3.

⁵⁹ Vitruvius Vl.iii.3: for the use of geometry to determine measurements on site, see IX.praef. 7.

⁶⁰ Compasses (with other drawing equipment) from Pompeii are described and illustrated in Dickinson, H.W., ‘A brief history of draughtsmen's instruments,’ Trans. Newcomen. Soc. xxvii (1927-1928), 73–4 and pi. xvi fig. 1: a pair of compasses was found in a house which had apparently belonged to an architect (M. della Corte, Case ed abitanti di Pompei (3rd edn, 1965), 164).Google Scholar

⁶¹ MacDonald, op. cit. (note 14), 168.

⁶² Vitruvius I.vi.6; IX.vii.3-6: V.vii.2.

⁶³ Vitruvius IX.praef. 6.

⁶⁴ Faventinus 28.

⁶⁵ See Chapman, H., ‘A Roman mitre and try-square from Canterbury,’ Antiq. Journ. lix (1979), 403–7. None of the examples given has a side of greater than about a foot, but these are probably carpenters' tools.Google Scholar

⁶⁶ Blagg, T.F.C., ‘Tools and techniques of the Roman stonemason in Britain,’ Britannia vii (1976), 170.Google Scholar

⁶⁷ Vitruvius l.vi.6; III.iv.5; VII.i.4: Faventinus 19.

⁶⁸ Vitruvius VIII.v.I. This passage, outlining levelling methods used in the construction of aqueducts, also mentions two other forms of level, the dioplra and the chorobates.

⁶⁹ Cunliffe, B., Excavations at Fishbourne 1 (1971), 78.Google Scholar

⁷⁰ Vitruvius VI.8.5-7.

⁷¹ Evans, E.M., Dowdell, G. and Thomas, H. J., ‘A third-century maritime establishment at Cold Knap, Barry, S Glamorgan,’ Britannia xvi (1985), 75, 80, 85.Google Scholar

⁷² The necessary theoretical knowledge existed: T. L. Heath, A Manual of Greek Mathematics (1931), 32ff.

⁷³ Breeze and Dobson, op. cit. (note 12), 56.

⁷⁴ R.E.M. and Wheeler, T.V., ‘The Roman amphitheatre at Caerleon, Monmouthshire,’ Archaeologia lxxviii (1928), 215.Google Scholar

⁷⁵ G.C. Boon, Isca (1972), 93.

⁷⁶ RIB 339, 342-5, 347. 350.

⁷⁷ Richmond, I.A., ‘Trajan's army on Trajan's Column,’ PBSR xiii (1935), 22f.Google Scholar

⁷⁸ I.A. Richmond, Hod Hill H: Excavations between 1951 and 1958 (1968), 74.

⁷⁹ Hanson, W.S. et al. , ‘The Agricolan supply-base at Red House, Corbridge,’ Arch. Ael. ⁵ vii (1979), 82.Google Scholar

⁸⁰ Richmond, loc. cit (note 78).

⁸¹ J.J. Coulton, Creek Architects at Work (1977).

⁸² See Appendix, p. 163.

⁸³ op.cit. (note 81), 68.

⁸⁴ Greek temple design, 85.

⁸⁵ Walthew, op. cit. (note 1).

⁸⁶ This is the practice recommended by Vitruvius in IV.iii.3 and VI.iii.4-5.

⁸⁷ This need not have posed any problems during the laying out of the building. All that is required is to take a piece of cord equal in length to the baseline, fold it into the number of parts required and mark them by attaching tags.

⁸⁸ Another building for which an analysis of the plan was included as part of the published report was the amphitheatre at Caerleon: for details of this, see above.

⁸⁹ A further room (Q) may also have belonged to the original layout, but this is difficult to prove.

⁹⁰ By J. Parkhouse. I am grateful to him and to the Glamorgan-Gwent Archaeological Trust for allowing me to use the data on this building and the accompanying plan. The structure will be fully published in J. Parkhouse and E.M. Evans (eds), Excavations in Cowbridge 1977-89 (forthcoming).

⁹¹ And Q is square (10 ft by 10 ft or 4 by 4 modules).

⁹² Zienkiewicz, op. cit. (note 15), 97-114.

⁹³ Greek temple design, 69.

⁹⁴ Walthew, op. cit. (note 1), 24-31.

⁹⁵ The south (front) wall has been drawn through the one post the position of which could be located.

⁹⁶ Pitts and St Joseph, op. cit. (note 33), 80.

⁹⁷ The other example cited by Walthew, Valkenburg Praetorium (his fig. 5) could more logically be analysed using a 5-ft modulus; although this does not entirely explain the plan which is markedly irregular, with the rooms on the northeast and southwest sides of the courtyard apparently laid off from different baselines.

⁹⁸ Richmond, op. cit. (note 78), 78.

⁹⁹ S.N. Miller, The Roman Fort at Balmuildy (1926), 22 and fig. 7; McDonald, A., ‘A note on the Roman fort at Croy Hill,’ PSAS xi (1936-1937), 37, pl. iii. Internal measurements given by McDonald do not include wall thickness.Google Scholar

¹⁰⁰ Maxwell, G., ‘Excavations at the Roman fort at Crawford, Lanarkshire,’ PSAS civ (1971-1972), 170, fig. 12.Google Scholar

¹⁰¹ Building inscriptions (RIB 2191-2 and 210-3) reveal the presence of Legion II Augusta at Balmuildy and a detachment of Legion VI at Croy Hill, so this would not appear to be a plan type linked with a specific legion.

¹⁰² Davison, op. cit. (note 34), 7, tables VI, XVIII, XXX (pp.599-607, 758-81, 894-7).

¹⁰³ ibid., 211, Walthew, op. cit. (note 1), 22-3.

¹⁰⁴ Gentry, A., Roman Military Stone-built Granaries in Britain, BAR 32 (1972), 7 and Appendix 4. (The data in Gentry's table suggest proportions between 2:5 and 2:9 might be nearer the truth.)Google Scholar

¹⁰⁵ Vitruvius 1.1.4.

¹⁰⁶ Vitruvius I.ii.2.

¹⁰⁷ CIL VI.29845. reproduced in G. Carettoni et al., La pianta marmorea di Roma antica (1960), 209 and F. Sear, Roman Architecture (1982), 69.

¹⁰⁸ Hudson, C., ‘Piante iconografiche incise in marmo,’ Roemische Mitteilungen v (1890), 48.Google Scholar

¹⁰⁹ op. cit. (note 107), 210.

¹¹⁰ Suetonius, Div. lul. 31.

¹¹¹ Aulus Gellius, Nodes Atticae XIX.10.1-3.

¹¹² Cassius Dio LXIX.4. The sense of this passage demands an elevation drawing or artist's impression of the interior, rather than a plan.

¹¹³ Pliny, Ep. IX.39.

¹¹⁴ A. N. Sherwin-White, The Letters of Pliny (1966), 52f.

¹¹⁵ See for example Vitruvius, Books II and IV.

¹¹⁶ ad Q. fr. III. 1.

¹¹⁷ JRS lvii (1968), 183Google Scholar; Britannia xxiii (1992), 281.Google Scholar