Hostname: page-component-586b7cd67f-t8hqh Total loading time: 0 Render date: 2024-11-22T04:35:18.988Z Has data issue: false hasContentIssue false
  • English
  • Français

Length-Units in House-Planning at Silchester and Caerwent

Published online by Cambridge University Press:  09 November 2011

C.V. Walthew
C.V. Walthew
Affiliation:
Department of Classics, University College, Dublin.
Get access Rights & Permissions [Opens in a new window]

Extract

In recent years there has been much discussion of the ancient units of measurement used in both civil and military planning in Roman Britain, without great consensus being achieved. Clearly, this is not a subject for dogmatic assertion, but prolonged re-examination of the evidence has convinced the writer that the units of measurement employed in planning the town houses of Silchester and Caerwent were 7 1/2 and 3 3/4 pedes Monetales. It will be argued that the 7 1/2 and 3 3/4 p.M. units (hereafter referred to as units and half-units) were used not only to lay out the houses themselves, but also to calculate their positions within their respective insulae and, indeed, the overall sizes of the insulae themselves. If right, these suggestions clearly have far-reaching implications for the whole subject of Roman urban planning.

Type
Articles
Information
Britannia , Volume 18 , November 1987 , pp. 201 - 231
Copyright
Copyright © C.V. Walthew 1987. Exclusive Licence to Publish: The Society for the Promotion of Roman Studies

Access options

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

References

2 Frere, S.S., Beiheft zum Bericht der Römisch-Germanischen Kommission lviii (1977), 87–103Google Scholar; Walthew, C.V., Britannia ix (1978), 335–50CrossRefGoogle Scholar; Duncan-Jones, R.P., Britannia xi (1980), 127–33CrossRefGoogle Scholar; Walthew, C.V., Britannia xii (1981), 15–35CrossRefGoogle Scholar; Crummy, P., Britannia xiii (1982), 125–34CrossRefGoogle Scholar; Millett, M., Britannia xiii (1982), 315–20.CrossRefGoogle Scholar

3 Except on the difficulties involved, which no-one has denied. It may be pointed out that this paper is not concerned with distinguishing between p.M. and p.D.

4 Huggins, P., Rodwell, K. and Rodwell, W. in Drury, P.J. (ed.), Structural Reconstruction, BAR 110 (1982), 21–65 (esp. p. 53).Google Scholar

5 For 7½ and 3¾ p. M. units and their Imperial equivalents see Table I.

6 See note 25.

7 This measurement was taken between the faces of the side walls and made no allowance for their thicknesses.

8 The high margins are clearly explicable in terms of the 3 ft. 10 ins. overall thickness of the outer walls.

9 Owing to the thickness of the outer walls, the ambiguity between 4 and 4½ units cannot be resolved. The east wall of the building was 1 ft. 8 ins. wide and the west unusually broad at 2 ft. 3 ins. (perhaps because it lay on the western edge of the insula). On balance, 4 units seem preferable and have been adopted in the overall calculation of the south frontage (see p. 216).

10 The margin may seem high, but the 4 ft. overall thickness of the outer walls should be noted.

11 The high margin is due to the wall-thicknesses on which no information was published.

12 Note the 3 ft. 8 ins. overall thickness of the outer walls.

13 The thickness of the west wall was not established.

14 The high margins are due to the 3 ft. 10 ins. overall thickness of the outer walls.

15 The published plan suggests 66 ft. rather than 68 ft., reducing the margin to + 1.15%.

16 Archaeologia lii,2 (1890), 735–40 and pl. xxviii; liii, 1 (1892), pl. xxi.

17 For the detailed breakdown of room measurements see FIG. IA.

18 Archaeologia liii, i (1892), 275–80 and pls. xxii, xxiii; lv, i (1896), 218–9 and pl. x, lv, 2 (1897), 418–20 and pl. xxiii.

19 Exceeding the room totals by half a unit in both cases, the discrepancy, as so often, being explained by the wall-thicknesses.

20 102 and 88 ft. are the measurements provided by the report.

21 Comprising 17 ft. for the internal width of the south wing, plus 6 ft. for the overall thickness of the three walls, plus 6 ft. for the east corridor: 29 ft. in total.

22 Comprising 17 ft. 9 ins. for the internal width of Rooms 1 to 4, 6 ft. for the thickness of the three NS walls and 7 ft. 6 ins. for the width of the west corridor, giving a total of 31 ft. 3 ins. Again, the margin of error might seem high (+ 7.76%), were the wall-thicknesses not allowed for.

23 Archaeologia liv, 1 (1894), 211–9 and pls. xvii, xviii.

24 Boon, G.C., Silchester: The Roman Town of Calleva (1974), 193.Google Scholar

25 Comprising 14 ft. 10 ins. (central range) + 3 ft. 8 ins. (inner walls) + 6 ft. 8 ins. (west corridor) + 7 ft. (east corridor) + 3 ft. (outer walls of corridors) for a total of 35 ft. 2 ins. The excavators give 36 ft., though this still falls short of 5 units. On these figures the margins of error may be thought unacceptably high (see Table 2. p. 203, + 7.8% and + 10.3% respectively, but if the overall thickness (3 ft.) of the outer walls is taken into consideration, then the correspondence to 4V2 units (3¾ ft. 7½ ins.) looks more convincing.

26 As suggested, for example, by the earlier lengthy east-west wall running beneath the northern end of XIV,i (FIG. 2A).

27 Archaeologia lv, 1 (1896), 219–34 and Pls. x to xrv.

28 ibid., 234–46 and pl. xv.

29 129 ft. is the figure given.

30 Archaeologia liv, 2 (1895), 440–1 and pl. xlv.

31 Archaeologia lvii, 2 (1901), 230–2 and pl. xxx.

32 Archaeologia lvii, 2 (1908), 204–6 and pl. xxiii.

33 Archaeologia lvii, 2 (1901), 238–41 and pl. xxx.

34 Compare Silchester IX, 2 (12½ units deep NS, incl. NE wing); XVI, I (N wing, Rooms 5 to 10, 12½ units EW); XVI,3 (12½ units NS); XXXIV, 1 (W frontage 12½ units NS); III, 5 (E frontage 12½ units NS); XXXIIIA, 5 (W frontage 12½ units NS); XXXIV, 1 (enclosure S of house 12½ units EW); XXXV.I (12½ units NS). Caerwent III, s (12½ units NS); IV, S (12½ units NS); V, S (N section 12½ units NS); VII, S (enclosure to SE 12½ by 12½ units); XI, S (12½ units EW); XV, S (E frontage 12½ units to SE corner of Room 18); XVI, S (N frontage 12½ units EW in Period 4); XXI, S (later N frontage 12½ units EW); III, N (12½ units long NS); VIII, N (original NS length 12½ units); XIII, N (S frontage 12½ units EW).

35 Room 8 in the east wing and Corridor 9 in the west wing both measured 29 ft. NS internally.

36 Archaeologia Iv, 2 (1897), 415–8 and pl. xxiii.

37 Archaeologia lvii, 2 (1901), 237–8 and pl. xxx.

38 5½ units figured prominently in the planning of other Silchester houses, e.g. VI,2, original west wing (FIG. 7B); XIV,2, east wing (FIG. 2A) and XIX,2, north wing (FIG. 5A). Significantly, all three were wings measuring 11 by 5½ units, that of XIV, 2 being later extended to the north by a further 5½ units. Compare also XXI, 2 (FIG. 10A) and XXXV, 1 (FIG. 4A). The 11 units long N wing of the latter falls into two halves each of 5½ units.

39 Archaeologia lvi, 2 (1899), 230–7 and pl. xi.

40 Since the narrower foundations of a timber-framed house might allow more precise calculations, it is worth noting that the rooms, 20 ft. wide internally, were flanked to the W by a wall 1 ft. plus thick and to the E by a wall 8 ins. thick, which produces a total convincingly close to 3 units (21 ft. 9 ins.). To this we may add the 7 ft. wide east corridor and its c. i ft. wide outer wall for an overall width of c. 29 ft. 8 ins.

41 It is worth noting the symmetrical pattern of room divisions in the west wing. From north to south the distribution of units is 2½ + 3 + 3(1 + 2) + 3½ + 2½. In the north wing we may compare the suggested 4 × 2½ units layout of Room 5 with the internal measurements of 28 by 18 ft.

42 Archaeologia lvii, 2 (1901), 301–10 and pl. xl.

43 See note 34.

44 Compare the 42 ft. 8 ins. internal width of the peristyle and the 14 ft. 6 ins. width (NS) of the rooms in the north wing.

45 Walthew, C.V., Britannia vi (1975), 191Google Scholar, 194; idem, Britannia ix (1978), 337–8.

46 Archaeologia lxii,2 (1911), 427–34, pl. lx and figs. 11–14.

47 The internal dimensions of the eastern block were 69 ft. 6 ins. NS by 23 ft. 3 ins. EW.

48 See note 34.

49 Archaeologia lxii,2 (1911), 421–7 and pl. lx.

50 Boon, G.C., ‘The Shrine of the Head, Caerwent’ in Boon, G.C. and Lewis, J.M. (eds.), Welsh Antiquity (1976), 169–70Google Scholar and fig. 3.

51 Archaeologia IXii, I (1910), 7–11 and pl. i.

52 Archaeologia lxii, 2 (1911), 415–7 and pl. lvii; lxiv (1913), 441–5 and figs. 1–4. XXIV, N. Period 1 (Rooms 6–9): 7 units EW by 3½ units NS. Period 2 (Rooms 4, 5, 10–12): 5½ units EW by 8 units NS. Period 3 (Rooms 1–3): 11 Vz units EW by 4½ units NS. The total amount of street frontage occupied by XXIV,N measures 11 units EW by 11 Vi units NS. XXV,S. Period 1: 3½ units EW by 3 units NS (25 by 18 ft. internally). Period 2: 15 units EW (no ft. given) by 4½ units NS. Period 3: 15 units EW by 6 units NS.

53 Archaeologia Ix, 2 (1907), 451–64 and pl. xlii.

54 Archaeologia lix, I (1904), 106–7 and pl. x.

55 Internal measurements given include 14, 14½, and 14½ ft.

56 Archaeologia liv, 2 (1895), pl. xlv. For the detailed measurements see Table 2, p. 203.

57 Archaeologia lv, 2 (1897), pl. xxiii. The C. 300 ft. EW measurement given for the insula is clearly incorrect.

58 Archaeologia liv, 2 (1895), pl. xlvi.

59 These are, of course, external dimensions. Block I had walls 2 ft. thick, Block III walls 1 ft. 10 ins. thick and Block V measured 23 ft. 10 ins. wide internally (see Table 2 p. 203).

60 Archaeologia lx, 1 (1906), pl. xxi. The stated dimensions of the insula are 231 ft. NS by c. 270 ft. EW.

61 Archaeologia lv, 1 (1896), pl. x. Stated dimensions 270 ft. EW by 236 ft. NS.

62 Archaeologia liii, i (1892), pl. xxi. Boon, op. cit. (note 24) gives the overall dimensions as 385 by 388 ft. (117.34 by 118.26 m.).

63 Archaeologia liv.i (1894), pl. xvi. Stated EW dimension of insula c. 385 ft.

64 For confirmation of the 52V2 units note the minor east-west street, 2 units wide, lying 18 units from the northern and 32½ units from the southern boundary of the insula and Boon's figure of 115.82 m/380 ft. for the overall north-south length of the insula: op. cit. (note 24), 95.

65 Archaeologia lviii, 1 (1902), pl. ii. Stated length 383 ft.

66 Archaeologia lx, 2 (1907), pl. xl.

67 For other 32 unit frontages see p. 226.

68 Archaeologia lviii,1 (1902), pl. viii. The 118 ft. (NS) by 250 ft. 6 ins. (EW) overall dimensions stated by the excavators for II,S do not appear to be very accurate. The published plan, on the other hand, seems to be highly accurate and, together with the internal room measurements listed (p. 121), permits precise calculation of units and half-units. Thus it may be seen that the E frontage of the house is no more than 109 ft. (NS) and that the EW dimension up to the W wall of Room 25 (second NS wall from the left on FIG. 2D) does not exceed 246 ft. When converted into units (15 and 34 respectively) these figures correspond closely with the subdivisions proposed on FIG. 2D.

69 Archaeologia lviii, 2 (1903), pl. xxvi.

70 Archaeologia lxii, 2 (1911), 420–41, pl. lx and fig. 18.

71 See p. 214.

72 On VII and VIII, S see Archaeologia lviii, 1 (1902), 138–51 and pl. viii.

73 Archaeologia lix, i (1904), pls. ix and x; ix, 1 (1906), pl. xvii; lxi,2 (1909), pl. xci.

74 Archaeologia lxii, 2 (1911), pl. lvii.

75 Archaeologia lx, 2 (1907), pl. xlii; lxii, 1 (1910), pl. 1. The temple precinct measured no ft. NS by 63 ft. EW.

76 Nash-Williams, V.E., Bull. Board Celtic Stud. xv (1954), 159.Google Scholar

77 Cf. the 44½ unit maximum NS length occupied by buildings in the insula containing the supposed inn (House XII,S): Archaeologia lix, 2 (1905), pl. lxvi.

78 Archaeologia liii, 1 (1892), pl. xxii; lx, i (1906), pl. xxi. The NS dimension given for Insula V is 235 ft.

79 Cf. the two-roomed building on the N frontage of Insula XXXIV standing 24 units from the NE corner (FIG. 10B).

80 Archaeologia lvii, 2 (1901), pl. xxx. The c. 394 ft. width.of the insula seems to correspond well enough with the 54 units calculated for its S frontage.

81 Archaeologia liv, 2 (1895), 444–8 and pl. xlv.

82 Archaeologia lvii, 1 (1900), 98–9 and pl. ix.

83 Archaeologia lix, 2 (1905), 339–40 and pl. lxxiii.

84 Archaeologia lviii, 1 (1902), pl. viii.

85 Archaeologia liii, i (1892), pl. xxii.

86 Archaeologia lx, 1 (1906), pl. xxi.

87 Archaeologia viii, 1 (1900), pl. viii. The c. 260 ft. stated as the EW dimension of the insula is patently incorrect; the plan indicates c 267 ft.

88 On the timber-framed building see Boon, op. cit. (note 24), 198–9.

89 Another possible instance of the observation of a demarcation-line is perhaps to be seen in Insula XIX (FIG. 5A), where XIX,1 stands 16½ units from the NE corner and the main part of XIX,2 extends no further than 16½ units W of the E frontage.

90 Archaeologia lxiv (1913), 445, note 2.

91 Archaeologia lix, 1 (1904), pl. ix.

92 Archaeologia lix, 2 (1905), pl. lxvi.

93 13½ units would seem to represent the original widths (EW) of both VII and VIII,S. (Archaeologia lviii, 1, 1902, pl. viii) and 14½ units the original width of XV,S (see p. 214 and FIG 7A).

94 This rests on the assumption that the Period 3 east wall of the yard to the E of VII,N follows an earlier boundary line.