the ancient state prison of Rome, situated
between the temple
of Concord and the curia at the foot of the Capitol (Liv. i.
urbe imminens foro
). Cf. Veil. ii. 7
. 2; Val. Max. ix. 12
6; Plin. vii.
; Seneca, controv. ix. 27. 20; Fest. 264.
It was used simply as a place of detention, and not of
though executions (i.e. those of Jugurtha and
Vercingetorix and of the
Catilinarian conspirators) also took place here. The
was called Tullianum (applied in Amm. Marc. xxviii. 1
57 to the whole).
The name (Liv. xxix. 22
. 10; xxxiv. 44
. 8; Serv. ad Aen.
Flacc. decl. 4: Acta Chrysanth. et Dariae, 25 Oct. p. 483)
is by Varro
(LL v. 151
) and Festus (356) derived from Servius
Tullius, who was
the builder of this portion of the carcer : while Livy (i.
33) attributes the
construction of the carcer to Ancus Martius. Sallust
(Cat. 55) describes
it in a well-known passage: in carcere locus quod
ubi paullulum ascenderis ad laevam, circa duodecim
pedes humi depressus. Eum muniunt undique parietes atque insuper
fornicibus iuncta, sed incultu, tenebris, odore foeda atque
This lower chamber of the building is subterranean and
accessible only by a hole in the roof. It is nearly 7
metres in diameter:
in the walls only three courses of stone are visible, and it
is thus less
than 6 feet high: but three more courses may still be
hidden by the
present floor, and this would give the 12 feet of which
The building was, according to one theory, in origin a
cupola grave, like
those of Mycenae: while others think that it served as a
and derive the name Tullianum from tullus, a spring. A
does indeed still rise in the floor; and the absence of
as an argument against the second hypothesis, has little
weight, as the
water is not calcareous.
It has generally been believed that the cupola was cut
by the constructors of the upper chamber; Tenney Frank (TF 39-
47) now supposes,
without sufficient reason, that the lower chamber
originally had a flat
wooden roof, which later served as a scaffolding for the
flat stone vault,
which dates from after 100 B.C. But the holes to which
he points in
support of this theory may just as well have been cut for
There is little doubt that the chamber was originally
circular (the statement that the straight chord on the side towards the
Comitium is of rock,
is incorrect). See JRS 1925, 121
Most authorities attribute to it a high antiquity: but
the lower chamber to the third century B.C. owing to the
use of peperino
(not tufa, as all other authorities state) and the regularity
of the blocks,
uniformly 56 cm. high: while the date of the drain leading
forum appears to be debateable.
The upper room is a vaulted trapezoid, the sides
varying in length
from 5 to 3.60 metres. This Frank assigns to about 100
B.C. on similar
grounds; and the vault of the lower chamber, as we have
seen, to a
slightly later date.
A new facade of travertine was added by C. Vibius
Rufinus and M.
Cocceius Nerva, consules suffecti, perhaps in 22 A.D.
cf. 9005; Pros. i. p. 428, No. 972; iii. p. 424, No. 395),
but, it may be,
a good deal later (Mommsen, Westdeutsch. Zeitschr.,
blatt, 1888, 58, puts it a little before 45 A.D. ; cf. ILS iii.
It was still used as a prison in 368 A.D. (Amm. Marc.
, 57), so
that the tradition that it was converted into an oratory in
century is without foundation; and the fons S. Petri, ubi
est carcer eius
of Eins. (7. 2), cannot have been here (Mon. L. i. 481
The name Mamertinus is post-classical.
The building near the Regia, mis-called Carcer by Boni,
is a series of
and may belong to about 70-40 B.C. (CR 1902,
; Mitt. 1902,
; 1905, 116-117
; TF 87; HC).
See Jord. i. 2
. 323-328; RL 1902, 226-239
; HC 119-
; ZA 60-63; Leclercq in Cabrol, Dict. v. 2053
2057; HFP 5-8.