previous next


the baths erected by Diocletian on the high ground to the north-east of the Viminal. The dedicatory inscription (CIL vi. I 130=31242) of which at least four copies were set up runs as follows:D(omini) n(ostri) Diocletianus et Maximianus invicti seniores Aug(usti), patres imp(eratorum) et Caes(arum), et d(omini) n(ostri) Constantius et Maximianus invicti Aug(usti), et Severus et Maximinus nobilissimi Caesares ... thermas felices Diocletianas, quas Maximianus Aug. rediens ex Africa sub praesentia maiestatis disposuit ac fieri iussit et Diocletiani Aug. fratris sui nomine consecravit, coemptis aedeficiis pro tanti operis magnitudine omni cultu perfectas Romanis suis dedicaverunt. Maximian's return to Rome took place in the autumn of 298; while the abdication of Diocletian and Maximian in favour of Constantius (Chlorus) and (Galerius) Maximianus took place on Ist May, 305, and Constantius died on 25th July, 306. The inscription therefore belongs to the period between the two latter dates, and the baths took between seven and eight years to complete. It is noteworthy that the bricks used belong entirely to the period of Diocletian, no older material having been employed. The exterior (like that of the thermae of Caracalla and of the curia) was faced with white stucco in imitation of construction in blocks of white marble. The date given by Hier. a. Abr. 2318 (302 A.D.), cf. Chron. 148, is therefore incorrect. The baths are also mentioned in Not. (Reg. VI) and in Hist. Aug. (xxx. Tyr. 21), where we are told that the QUADRIGAE PISONIS (q.v.) were among the buildings removed to make way for the baths. We may add a monument of an undetermined period,1 decorated with sculptures, including a relief representing the temple of Quirinus (Mitt. 1904, 23-37), and various private houses, including that of CORNELIA L. F. VOLUSI SATURNINI (q.v.); while the north-eastern portion of the vicus Longus was suppressed, and the Alta Semita and the Vicus Collis Viminalis connected by a new cross street (BC 1880, 132; 1887, 181; 1888, 36; the streets are shown by LF 17 and HF i, ii; cf. RhM 1894, 383, 388).

The statement in Hist. Aug. Prob. c. 2: usus autem sum praecipue libris ex bibliotheca Ulpia, aetate mea thermis Diocletiani, is a pure invention according to v. Domaszewski (SHA 1916, 7. A, 9). If it is true that these thermae could accommodate 3000 people in marble seats, almost double the number of bathers that found room in those of Caracalla (Olympiod. ap. Phot. 80, p. 63 a, Bekker, v. p. 521), then, inasmuch as the area is about the same, the space must have been more economically used. They are also mentioned by Schol. Iuv. xi. 56, and Sidon. ad Consent. 495. CIL vi. 1131 (the date of which is uncertain) refers apparently to a restoration, saying 'thermas Diocletianas a veteribus principibus institutas omn[i cultu ... restituit].' The destruction of the aqueducts in the Gothic wars naturally rendered them unusable; but they are mentioned as in the fourth ecclesiastical region in Regest. Honor. i. a. 625 ap. Deusdedit (iii. c. 138); and the name in thermis Diocletianis was applied to the church of S. Cyriacus right through the Middle Ages (HCh 245, 246), while in the Mirabilia (27) and in Magister Gregorius (JRS 1919, 52) the building is known as the Palatium Diocletiani.

The thermae of Diocletian (Ill. 54, a view taken shortly after the construction of the first railway station) occupied about the same area as those of Caracalla (a rectangle of about 356 by 316 metres, or about 281 acres) and closely resemble them in plan. The central hall of the main building, which measured 280 by 160 metres, wrongly known as the tepidarium until quite recently, is derived, as Rivoira points out, from that of the thermae of Titus and of Trajan, and is very similar to that of the baths of Caracalla; while from it is derived the plan of the Basilica of Constantine. Its excellent preservation is due to its conversion into the church of S. Maria degli Angeli by Michelangelo; though there is not sufficient evidence to allow us to attribute to him, instead of to Vanvitelli, the new apse on the north-east side (cf. Roma iii. (1925), 349- 356, 395-408). It has an intersecting vault divided into three bays; the four columns of grey granite on each side do not support the vault, but are purely ornamental. The four smaller rooms at the angles may havc served for cold baths, as there is no trace of heating; while between them, on the minor axis, there was access to the frigidarium on the north-east and to the circular tepidarium (now the vestibule of the church) and the rectangular caldarium, which projected south- westwards, and though extant in the sixteenth century is now destroyed; see DuP 127.

On the major axis, on the south-west, there was an approach at each end through two rectangular halls (on each side of which were others) to the palaestrae, one at each end of the main block on each side of the frigidarium, a hall containing a huge shallow bathing pool, which was open to the air; its north wall, elaborately decorated with niches, is still in great part preserved; see Piranesi, Vedute di Roma, No. 115 Hind, and 111. 53. On each side of the caldarium were the apodyteria or dressing rooms, and other halls, which served for private baths, etc., as well as for conversation, recitations of poets, rhetoricians, etc., and completed the rectangular central block.2 This was surrounded by a garden, which was enclosed by an outer peribolus. Around this were small rectangular halls and semicircular exedrae, which were also used as reading and lecture rooms, gymnasia and lounging rooms.

In the centre of the south-west side was a very large exedra, which was doubtless provided with seats and served as a theatre (like the corresponding exedrae in the thermae of Trajan and Caracalla).

The actual enclosure wall was preserved until modern days. The Via Nazionale was driven through it in 1867, and only the line of its curve is still shown by the buildings of the Piazza dell' Esedra dei Termini, the corrupt form in which the name of the thermae still lingers on. At the west and south angles of the peribolus are two circular halls, one of which is especially well preserved, owing to its conversion into the church of S. Bernardo in 1598.3

The whole of the external brick facing was covered with plaster, in imitation of construction in blocks of white marble with draughted joints: this was also done in the thermae of Caracalla, the basilica of Constantine, etc.

The reservoir by which the baths were supplied was fed by the aqua Marcia, the volume of which was increased by Diocletian (see p. 27). It lay outside the peribolus on the south side; and, being in the angle between the baths and the vicus Collis Viminalis, it was trapezoidal in shape, 91 metres in length, with an average width of 16 metres. The last remains of it above ground were not destroyed until 1876 (Falda, Giardini di Roma, 14; LA 308, 318; BC 1872-3, 230; 1906, 106-107; Archaeologia li. 502, fig. 12; HJ 382, n. 22; JRS 1919, 190).

For the excavations and demolitions of the fifteenth and sixteenth centuries and the condition of the ruins at that period, see BA 1909, 361-372, 401-405; LS passim (esp. i. 148; ii. 135-149); DuP 125- 128; HCh 245, 246, 589, 590. The ruins of the caldarium and the southern angle of the central block were in large measure removed by Sixtus V.

For the thermae in general, see Sebastianus de Oya,4 1558 (Brunet, Manuel iv. 302) ; Paulin, Restauration des Thermes de Diocletien 5 (Paris 1890); D'Esp. Mon. ii. 162-171 ; Fr. i. 98, 99; LR 434-437; LS passim; HJ 377-382; BA 1911, 347-361 ; Paribeni, Le Terme di Diocleziano ed ii Museo Nazionale Romano, 9-56; Toeb. i. 1113; RA 204-210 (superseding Journal of the Brit. and Am. Arch. Soc. iv. 353-360); Mem. L. 5. xvii. 533; RE ii. 2755; ASA 106, 107.

1 For the date see p. 439, and n. I.

2 The identification of the rooms at the south-east and south-west angles as laconica (Mitt. 1920, 168) is incorrect.

3 See S. Ortolani, San Bernardo alle Terme (Chiese di Roma illustrate, No. 8; Rome, n.d.).

4 See Journal Royal Inst. Brit. Arch. xxxi. (1923-4), 153.

5 On Paulin's work see Mitt. 1892, 308-311; and for unpublished drawings in Berlin, see Jessen in Aus der Anomia (1890), 119-121.

hide Dates (automatically extracted)
Sort dates alphabetically, as they appear on the page, by frequency
Click on a date to search for it in this document.
302 AD (1)
hide Display Preferences
Greek Display:
Arabic Display:
View by Default:
Browse Bar: