With its detailed publication, the excavations in the
Armenian Garden under the direction of A. D. Tushingham27
from 1961 to 1967 provide lots of material for studying
the Byzantine–Early Islamic period. This site might also
show archaeological evidence for the AD 749 earthquake
which was not interpreted as such by the excavator.
Three phases of buildings of the Byzantine–Early Islamic
period were distinguished: the phase entitled ‘Byzantine I’,
which clearly served a religious (Christian) purpose, likely
a chapel or even a small church.
28 Attributed to this phase
are an apse, the ‘Hare Mosaic’, an additional mosaic, some
cisterns, and a few walls.
29 The excavators attribute this
phase to the end of the reign of Justinian. A. D. Tushingham
assumes this phase was short,
30 suggesting an “abrupt end
by a washout.”
31
After that, a completely different constructed building
sounded the bell for a new architectural phase in this area.
The so-called ‘Byzantine II’ phase consists of a building
erected with a completely different orientation. It had a
central courtyard and three adjoining smaller rooms. A. D.
Tushingham describes this building as a possible khan or a
caravanserai.
32
Additionally, an area enclosed with walls containing structures
which are called ‘bins’, is attributed to this period.
33 Their
function is not entirely clear. The bins seem to have contained
several pieces of burned plaster and burned stones as well as a
large number of tiles.
34 However, possible use as a kiln was
excluded by the excavator.
35 This phase was brought to an
end by a ‘great washout’, similar to the previous one. It is
described by A. D. Tushingham as follows: “It is difficult to
believe that rains alone could have been responsible for such a
flood, but there is no doubt that the water was the agent of the
destruction.”
36
The last phase, ‘Byzantine III’, was a reconstruction of the
previous one but with a reduction of the large central courtyard,
which was cut in half by an inserted wall.
38 The excavator
interpreted both of the latter buildings as a possible
caravanserai or a pension for pilgrims.
39
In the light of the interpretation of the DEI area I results,
the transition between the different phases should be reassessed.
A. D. Tushingham claims that the last of the phases,
‘Byzantine III’, was destroyed during the seventh century AD,
either by the Persian conquest in AD 614 or by the ‘Muslim
conquest’ in AD 638 and only resettled in the medieval period.
40
Such an occupation gap does not seem very likely, considering
that during the Umayyad period, the south-western hill was still
a flourishing place, especially for Christian pilgrims, as is known
from various sources.
However, A. D. Tushingham still reports excavated material,
such as pottery and coins, from the Umayyad and Abbasid periods,
mainly in various fillings.
41 This does indicate that the area was
occupied during this period, but no architectural features were
attributed to this period.
J. Magness reassessed the pottery assemblages found in the
excavations of the Armenian Garden. She discovered that the
pottery of the Byzantine levels includes a mixture of types that
can be dated from the sixth to the eighth centuries AD. Therefore,
she concludes that it might be possible that the settlement phase
in the Armenian Garden cannot be dated exclusively to the
Byzantine period but can also have continued throughout the
Umayyad period.
42 Considering again the archaeological
evidence of continuous occupation from DEI area I,
43 a similar
inhabitation pattern throughout the Umayyad period in the nearby
Armenian Garden seems very likely.
Therefore, the author suggests in the following a reassessment
of the Byzantine occupational phases — at least the later ones —
into the Early Islamic period. Instead of ‘washouts’ (or the ‘Muslim
conquest’) which were described by the excavator, these events
might be connected to the earthquake. Following each ‘washout’
the area was rebuilt differently. Similarities can be drawn with the
excavation results of the nearby DEI area I — after the destructive
earthquake, the urban layout was altered drastically, thus changing
the nature of the site.
The transition between the excavator’s phases ‘Byzantine I’ and
‘Byzantine II’ in particular may be looked at here: from a chapel to an
architecturally completely different, possible caravanserai.
Two possible solutions can be considered for the reconstruction
of the Armenian Garden’s history:
Either the earthquake occurred
between Tushingham’s phases ‘Byzantine I’ and ‘Byzantine II’ and
as a consequence, the former chapel with an apse was overbuilt by
the suggested caravanserai (instead of rebuilding the chapel). In any
case, these measures changed the nature of the site altogether. The
other option is the modification of the caravanserai in Tushingham’s
phases ‘Byzantine II’ and ‘Byzantine III’ might be a possible result of
rebuilding measures following the earthquake.
44
J. Magness has also reassessed the dating of the repairing of the
city wall to the Early Islamic period (see below); as is analysed by A. D.
Tushingham, this repair work corresponds with his ‘Byzantine II’
period.
45 This dating would confirm that Tushingham’s ‘Byzantine II’
phase can be dated post-earthquake.
Consequently, the ‘Byzantine III’ phase was a time in which the
Armenian Garden was extensively repaired and rebuilt. Perhaps
this was necessary after a destructive event, such as an earthquake,
similar to DEI area I.
Furthermore, the so-called ‘bins’ mentioned
above, which can also be attributed to the possible early Islamic
layers, contain a lot of burnt material and former building material,
such as burnt plaster and tiles, as described by A. D. Tushingham.
47
It can be suggested that this material could be the remains of the
destroyed former buildings, the ‘Byzantine I’ structures. Those ‘bins’
were sealed, as described by A. D. Tushingham, by the latest
‘Byzantine III’ floor, which supports that argument.
48 For this, a
detailed find analysis of the content of these ‘bins’ would be necessary.
Certainly, this chapter cannot provide a complete detailed
reassessment of all the stratigraphical contexts and corresponding
finds of the excavation in the Armenian Garden; still, it can provide
cause to reconsider the archaeological history of the site during the
transitional period from the Byzantine to the Early Islamic period.
Footnotes
27 From 1962 onwards; the first year (1961) was under the
direction of R. de Vaux and J. A. Callaway (Tushingham 1985, 3).
28 Tushingham 1985, 101, pl. 6.
29 Tushingham 1985, 101, pl. 6.
30 Without providing a further time frame for “short.”
31 Tushingham 1985, 101. It remains uncertain what exactly
is meant by “washout.”
32 Tushingham 1985, 104.
33 Tushingham 1985, 101–02, pl. 6.
34 Tushingham 1985, 79.
35 Tushingham 1985, 79.
36 Tushingham 1985, 69.
37 Which is actually subdivided into IIIA
(redressing the damage caused by the previous 'washout') and IIIB
(actual rebuilding measures; Tushingham 1985, 103).
38 Tushingham 1985, 103–04.
38 Tushingham 1985. 103-04.
39 Tushingham 1985. 104.
40 Tushingham 1985, 104-05. Many earlier scholars,
especially from the 1960s and 1970s, attributed any
urban changes during this period to the `Muslim
conquest, neglecting the possibility of the AD 749
earthquake. But this view is no longer reflected in
modern research (Avni 2014, 14).
41 Tushingham 1985, 105-06. He wrote that the majority of
the Umayyad and Abbasid coins stems from 'deposits
that on stratigraphic grounds can be assigned to the
fills chat precede the first medieval, that is Ayyubid,
occupation of the site' (Tushingham 1985, 106).
42 Magness 1991, 212.
43 Zimni 2023; Namdar and others 2024.
44 We must also consider the possibility that several earthquakes
or aftershocks occurred in a short time, causing two different
“washouts” in the Armenian Garden.
45 Keeping in mind, that these repairs are either a result of the
earthquake or might also be the result of the reconstruction of the
city walls by Caliph Hisham.
46 Tushingham 1985, 65.
47 Tushingham 1985, 79.
48 Tushingham 1985, 79.
