JOURNAL OF WETLAND ARCHAEOLOGY 4, 2004, 139–153

That Sinking Feeling:

Wetland Investigations of the Origins of Venice

R. A. Housley, A. J. Ammerman and C. E. McClennen

ABSTRACT
It is well known that Venice has always had an intimate association with the sea. Despite
the historical interest the origins of the city have, until relatively recently, remained
cloaked in obscurity. Until the mid-1980s Venice remained one of the few major historical
cities in Europe where urban archaeology had yet to uncover significant information.
Over the last two decades the situation has changed as shown by fifteen sites with early
levels. Despite the difficulties of working in a tidal lagoon at considerable depths,
archaeology has begun to chart the earliest inhabitation. This paper outlines some of the
findings and the methodological approach adopted.

Keywords: EARLY MEDIEVAL, VENICE, LAGOON, WETLAND INVESTIGATIONS, SAN LORENZO DI CASTELLO,
SAN FRANCESCO DEL DESERTO, TORCELLO

Introduction
For many years Venice has been renowned for its long historical tradition and its artistic
heritage. The origins of the city and the early centuries are, however, much less known.
In chronicles written in the later Middle Ages the Venetians elaborated several legends
about how their city began. The historian today is, for the most part, quite sceptical about
the tradition that Venice was ‘born free and Christian’ (Brown 1991). From the 11–12th
century AD written and pictorial sources on Venice start to become more numerous
(Agazzi 1991; Crouzet-Pavan 1992; Schulz 1991). As a city built on water and with a
strong maritime history, one might think that Venice should have a long record of
archaeological wetland investigation, but this is not the case. In fact archaeological

Authors’ Addresses: Rupert Housley, Department of Archaeology, University of Glasgow, Gregory

Building, Lilybank Gardens, Glasgow G12 8QQ, UK; email: [email protected]
Albert J. Ammerman, Department of the Classics, Colgate University, 13 Oak Drive, Hamilton,
New York 13346–1398, USA.
Charles E. McClennen, Department of Geology, Colgate University, 13 Oak Drive, Hamilton,
New York 13346–1398, USA
140 R. A. HOUSLEY, A. J. AMMERMAN AND C. E. MCCLENNEN

excavation of the earlier levels of the city began only comparatively recently. Prior to
1985 there had been only one modern excavation that reached levels associated with the
early habitation of the city – in the vicinity of the Church of Santa Maria Assunta on the
island of Torcello (Leciejewicz et al. 1977; Tombolani 1988). However, subsequent
palaeoenvironmental studies that we have undertaken have shown the island of Torcello
to be much smaller in late Roman times than previously thought. Thus there is now the
realisation that the excavations by the Polish mission and Tombolani had taken place
off the original island in areas that were only subsequently reclaimed from the Lagoon.
Such problems, in conjunction with the difficulty of undertaking excavations in a dense
urban fabric with few open spaces and at depths well below mean modern sea level,
meant that little was known of the original habitation of the area.
However, in the last two decades this has begun to change, as archaeological and
palaeoenvironmental studies have taken place both ahead of, and in parallel with,
building restoration work. Since the mid 1980s archaeological fieldwork has led to the
discovery of fifteen sites with cultural horizons dating to the period before the 9th century
AD (Figure 1). This recent work, in combination with current studies of the wider lagoon
environment, has now begun to provide a better insight into the origins of the city.
Methodologically, it is the combination of a rich and complex urban archaeology, located
within a tidal saltwater setting that is undergoing regional subsidence, which makes for
some of the special aspects found in this study. The aim of this paper is to show how the
fieldwork is being undertaken in Venice through consideration of some of the more
significant new discoveries.

Aspects specific to the early archaeology of Venice

Ancient written sources and the earliest Venetian chronicles

Venice is virtually unique in northern Italy in being a major urban centre with no Roman
roots. Through the 3rd century AD, whenever an ancient source mentions Venetia, the
term is used with reference to a broad geographical region, and not a town. The earliest
source with any real detail relating to the Venice Lagoon is the well-known letter that
Cassiodorus wrote to the Lagoon dwellers in AD 537–81. There is no mention of Venice
being an urban centre prior to the writings of Paul the Deacon (History of the Lombards
II.14) in the late 8th century. The Venetian chronicles, which start with the account
written by John the Deacon in the first half of the 11th century,2 attempt to go back to
much earlier times but there are always nagging questions about their reliability. It is not
until the 12th century that the historical evidence becomes more comprehensive, and the
paucity of detailed information for the late Roman and early medieval periods means
that the onus of uncovering early Venice rests with the discipline of archaeology.

Tides and the Lagoon Environment

The Venetian Lagoon, at 550 km2, is a large, shallow body of water at the head of the
Adriatic separated from the sea by a narrow strip of sand, the barrier beach known as the
 THAT SINKING FEELING: WETLAND INVESTIGATIONS OF THE ORIGINS OF VENICE 141

Fig. 1. Map of the northern Venice Lagoon showing the location of the 15 sites where excavations
have uncovered archaeological remains that date from before the 9th century AD. (Based on
Ammerman and McClennen 2001, 14.)
142 R. A. HOUSLEY, A. J. AMMERMAN AND C. E. MCCLENNEN

Lido. At the height of the Last Glacial Maximum the Veneto region was a considerable
distance from the Adriatic shoreline, which lay some 275 km to the southeast beyond the
present day town of Ancona. Prior to the arrival of the sea the area comprised an extensive
river floodplain – part of a much-enlarged Po River valley. With climate warming and
melting of the continental glaciers, the sea level around the world began to rise rapidly.
The transgression of the marine waters rose vertically over 120 metres, reaching the
point where Venice is now to be found by the mid Holocene, i.e. around 6000 years ago.
Thus was formed the Venice Lagoon. Since then the Lagoon has been in continual
circulation with the sea through twice-daily tidal fluxes. This has created a highly
dynamic sedimentary environment with a continually changing mosaic of mudflats,
marsh islands and meandering channels. These three distinct sedimentary types
dominate the upper five metres or so of the stratigraphic record. Only in the upper levels
are in-situ archaeological structures and artefacts to be encountered. Some parts of the
Lagoon experience high-energy wave action and tidal currents, causing erosion were it
not countered by human intervention in the form of artificial public and private works.
In other localities the wave energy is weaker, leading to an accumulation of fine-grained
inorganic estuarine silts. Before the 9th century AD, the Lagoon would have offered rich
fishing, fowling and salt production opportunities, but except for reeds, and trees growing
along the banks of the tidal inlets, rivers and creeks, there would have been few resources
suitable for building a city.
Tides in Venice had, and continue to have, a profound influence on habitation patterns.
At the time of full and new moons, the difference in water level between high and low
tide on any given day would have been approximately 80–100 cm. In their building
practices the Venetians took account of this difference and also included an allowance
for the monthly and annual high-water levels. Thus pavements and living floors in both
early and later Venice were generally built at least 100 cm above the mean sea level
current at the time of construction. In other words, in order to avoid flooding Venetian
building practice was keyed to high water levels – locally termed acqua alta – rather than
to mean monthly and annual levels.
The extent to which it was feasible to move around the Lagoon in the past is a complex
question. As today, in the past it was possible to travel to most parts of the Lagoon by
small boat except at times of extreme low tide when large expanses of mudflat would
have been exposed. However, the water depth would have been too shallow to permit
the use of larger boats and sea-going ships except in the deeper channels. Study of the
prevalence and position of these deeper channels is complicated due to the processes
associated with lagoon tidal channel meandering (Figures 2 & 3). The deeper channels
are associated with higher tidal current speeds and thus tend to be sandier in composition
and are associated with dipping (sloping) channel point bar deposits. With continuous
twice-daily tidal flow reversals, the channels migrate over time, eroding and re-working
older sediments and creating scour and fill deposits. Due to the rapidity of accumulation,
plant material is often preserved in these point-bar deposits. While the shallower
channels simply re-work the upper lagoon salt marsh and mudflat sediments, the deeper
channels cut down into the pre-lagoonal floodplain strata that existed prior to the arrival
of the rising Adriatic Sea to the Veneto region.
This geomorphological framework has implications for the archaeology. Due to such
 THAT SINKING FEELING: WETLAND INVESTIGATIONS OF THE ORIGINS OF VENICE 143

Fig. 2. A stratigraphic cross-section of the Venetian Lagoon showing the main sedimentary units
and their approximate age. (McClennen, Ammerman and Schock 1997, 758.)

Fig. 3. Map of the area around San Francesco del Deserto showing the direction of movement
of the channel meanders within this part of the Lagoon. (CMcC)
144 R. A. HOUSLEY, A. J. AMMERMAN AND C. E. MCCLENNEN

sedimentary processes, it is unlikely that we will ever find an old ship or boat in the
bottom of a modern channel. For this reason underwater archaeology of this nature is
not likely to be productive3. Instead the place where early archaeological remains have
the best prospect of surviving, in a good state of preservation, is at a site with a long and
continuous history of occupation. In such situations the measures taken to prevent
erosion, and the progressive build-up of the ground level over the centuries, will have
enhanced preservation of the remains. The problem is in reaching such levels. One often
has to excavate down through architectural structures of considerable interest, which
limits the scope for archaeology.

Relative Sea Level and Subsidence in the Venice region

Many different factors can influence a change in the relative sea level of a particular
locality (Revelle 1990). They include: (1) changes in atmospheric pressure, winds or
ocean currents, (2) changes in the mass of ocean water brought about by wastage or
accumulation of global ice, (3) changes in the volume of ocean water due to temperature
fluctuations (excluding those attributed to water mass), and (4) earth movements such
as subsidence or uplift. As regards the last factor, in the case of Venice we are dealing
with subsidence rather than uplift, and the subsidence is influenced by a range of local
factors. These include tectonic movement linked to the thrusting of the Alps over the Po
Plain; the compaction of the sediments under loading; oxidation of the peat layers that
interleave with the riverine alluvium in the pre-Lagoon sequence; and groundwater
extraction for human and industrial use (Ammerman et al. 1999, 305). This last point is
more of relevance to present day Venice than to the past for it seems that groundwater
extraction between 1930 and 1970 was responsible for significantly increasing the rate
of subsidence in those years (ibid 306–7). When measuring changes in local relative sea
level it is important to have a common reference point. The tide gauge at Punta della
Salute is the best point of reference for measuring relative sea level because it has been in
continuous use for over one hundred years and is still the standard used today. For this
reason all the recorded elevations associated with our investigations are expressed
relative to this 1897 reference point. However, it must be recognised that regional
subsidence continues to affect the Punta della Salute reference point (Ammerman et al.
1999, 308).
Subsidence, taken in conjunction with the global eustatic rise in sea level over the last
few millennia and other factors like dewatering and loading, means that much of the
early archaeology of Venice is now well below present-day sea level. In practice in-situ
archaeological material may occur up to 2 m below the 1897 sea level reference point
(Ammerman et al. 1999, Figure 2; Ammerman in press, Figure 1). This means, in effect, a
depth of some 3 m below street level today. The logistic difficulty of undertaking
excavations in such situations helps to explain why the archaeology of early Venice got
off to a slow start.

Human impact and the transformation of the Lagoon

As will be seen below, there is good archaeological evidence for habitation of the islands
 THAT SINKING FEELING: WETLAND INVESTIGATIONS OF THE ORIGINS OF VENICE 145

within the northern Lagoon beginning in the first few centuries AD. Dry land suitable
for building purposes, as opposed to marsh islands prone to regular flooding, was at a
premium and so right from the start there was a strong incentive to alter conditions to
favour human needs. By raising the height of the land, through the importation of extra
sediment to a site, it was possible to reduce the probability of flooding in periods of acqua
alta; sand from some of the barrier island dunes seems to have been favoured at certain
times, or building rubble when it was available. Additionally, the enclosure and infilling
of areas of salt marsh which abutted the small natural islands on which early settlement
occurred allowed the inhabited area to expand. By building successive phases of wooden
waterside structures, to minimise erosion and increase the usable margins of islands, it
was possible to protect and maximise the land that already existed. Based on the
archaeological evidence now available (presented below) these methods were employed
in the Lagoon at least from Roman times.
As time passed, the scale of habitation increased and the nature of settlement changed.
Initially, habitation in Venice was more akin to a broadly dispersed series of settlements
than to an urban centre. Only later, in the 8–9th centuries AD, would it have taken on the
character of a city (Ammerman 2003). Maps of the 11–13th centuries, and detailed parish
records, suggest that even close to Piazza San Marco gardens, orchards and open places
existed. Certainly by the end of the 13th century Venice was a flourishing city with a
population of more than 100,000 inhabitants – the only other European centres of
comparable size at that time were Milan, Florence, Naples, Palermo and Paris. The city
grew using the age-old method of founding buildings on newly reclaimed land, on
wooden piles and boards. Later, brick and mortar became the favoured building medium
due to its comparative lightness.
On the extensive mainland behind the Lagoon, the region known today as the Veneto,
there was a system of rivers that drained into the Lagoon. By the 14th century human use
of this area for agriculture, and use of the rivers Brenta, Sile and Zero for transport,
meant a deterioration of the Lagoon, which began to silt up because of heavy sedimentary
loads. Shoals interfered with navigation, which threatened the trade-centred economic
basis of the city. If allowed to continue, the sedimentation would have opened Venice up
to direct military attack. Such a change would have been a disaster for the Venetians,
and so a start was made on schemes to divert the rivers away from the Lagoon so that
their sediment load would be discharged directly into the Adriatic. Thus the primary
source of alluvium responsible for causing shoaling in the port of Venice would be
eliminated, and the lagoonal salinity increased as less freshwater entered the Lagoon.
Over the next few centuries this was gradually achieved such that a map from 1709
shows how levees and dikes had totally diverted all the main rivers around the Lagoon
(Ammerman and McClennen 2001, 27). This illustrates how, in so many ways, historical
Venice is the consequence of human environmental interventions that, from small
beginnings, ultimately transformed the city and Lagoon to the situation visible today.

Approaches to Studying the Lagoon and its Archaeology

In the course of studying the early origins of Venice, three novel approaches proved to be
especially beneficial. The first was the initiation of a programme of hand coring on land
146 R. A. HOUSLEY, A. J. AMMERMAN AND C. E. MCCLENNEN

and also from small boats that could access all parts of the Lagoon. On first thought, the
latter might not seem particularly novel since previous knowledge of the Lagoon was
based upon an extensive series of cores used to investigate the stratigraphy and
depositional environments (Gatto 1980). However, it is now apparent that two limitations
affected this earlier work. Firstly, the cores were taken from work barges that could only
sample canals, channel edge sites and inlets where the water depth was sufficiently
deep to permit access of deep draught boats. Failure to sample the shallow mid-saltmarsh
island and mid-mudflats meant that the low energy environments were missed. The
second was the realisation that the sampled high-energy localities contained mostly re-
worked material and that the associated conventional 14C dates on marine shells had to
be biased. The large-sized marine shells selected for dating had to be sufficiently robust
to stand up to abrasion and were very likely re-worked. Furthermore, the difficulty of
correcting for different carbon sources in the shells – both the marine reservoir effect
(Stuiver et al. 1986, Stuiver and Braziunas 1993) and the dissolved carbonate in the river
water from the Dolomite region of the Italian Alps – meant the resulting ages were
potentially seriously misleading.
The second technique that proved to be crucial in improving our understanding of the
Lagoon and its archaeology was the development of the accelerator mass spectrometric
(AMS) 14C dating method and its application to small samples such as peat lenses and
wood and reed fragments that could be recovered from deep excavation soundings and
auger cores. Whenever possible we have tried to date at least ten samples from a range of
stratigraphic contexts on a given site, in order to establish a firm framework of absolute
chronology. Choice of materials that get their carbon from the atmosphere avoided the
problem of complex carbon sources associated with the marine shells and allowed single
entities to be dated from a wider range of localities. Furthermore, this approach to dating
overcame the difficulty prevalent in the upper levels of Venice of having to decide whether
or not culturally diagnostic artefacts were in-situ and thus indicative of the age of the
context in which they were found. Certainly the Venetian practice of using any available
material, including old building rubble, as landfill to build-up the land surface meant
that full confidence could not be placed in finds of such material as chronological
indicators.
The third methodology that greatly helped develop a much-improved understanding
of the sedimentary dynamics of the Lagoon was the use of high-resolution seismic-reflection
sub-bottom sonar profiling (McClennen et al. 1997). This involves the use of sonar echoes
to resolve contrasting sedimentary layers in the Lagoon. The sound waves that the equipment
generates can penetrate through the 5–6 m of lagoon sediment down into the fluvial and
peat horizons that were laid down in the Late Pleistocene period: 14C dates on the peat fall
in the 19,000–25,000 uncal years BP range (Figure 2; Ammerman et al. 1995). As the survey
vessel proceeds over the bottom, reflected echoes from the sonar display various sedimentary
features like horizontal beds, dipping layers and channel banks, as well as scour and fill
deposits. By continually recording the position of the vessel using a global positioning
system (GPS), it was possible to plot a record of the sub-bottom features. Part of the success
of this work again stems from the use of a small boat, which allowed us to range widely
over large areas of the Lagoon. In conjunction with hand coring, this methodology has
allowed a fundamental revision of the sedimentary framework of the Lagoon.
 THAT SINKING FEELING: WETLAND INVESTIGATIONS OF THE ORIGINS OF VENICE 147

The Archaeology of Early Venice

To illustrate the impact that the new forms of investigation have had on our under-
standing of the early origins of Venice, we shall now examine the findings of three
excavations undertaken in the 1990s. Firstly, the excavations beneath San Lorenzo di
Castello, a 9th century church extensively remodelled in the 16th century and lying
within the confines of the present city; secondly the site of San Francesco del Deserto, a
Franciscan monastery on an island in the northern part of the Lagoon; and thirdly
Torcello, the site of the former cathedral established in AD 639 when the Bishop of
Altinum relocated his seat from the mainland to this island in the north of the Lagoon.

San Lorenzo di Castello

Standing midway between San Marco and the Arsenal is the church of San Lorenzo di
Castello that was completely rebuilt at the end of the 16th century to the design of Simon
Sorella. For some time the present structure had suffered stability problems, to the extent
that in 1987 urgent restoration work had to be carried out. This provided the opportunity
to make a series of archaeological soundings to learn about the earlier structures that
occupied the site. A well preserved floor mosaic dating to the 12th century was uncovered,
which overlaid an apse supported by sandstone foundation blocks that, in turn, rested
on vertical wooden posts the tops of which were at an elevation of –0.8 m relative to
mean modern sea level (De Min 1990). An AMS date on the outer ring of one of the piles
subsequently showed that they probably dated to the 9th or first half of the 10th century
(Ammerman et al. 1992), which ties in well with the earliest literary record for the area
that informs us that the first church at San Lorenzo was built by the Badoer family at the
beginning of 9th century AD (De Min 2000). But this was not the earliest human activity
on the site. At lower elevation (around –1.9 m with respect to mean modern sea level)
were more cultural remains in the form of inorganic building debris (brick, tile, mortar,
slag and pottery of an undiagnostic form) and organic matter (wood fragments and a
matted reed structure). Hand auguring demonstrated that this cultural horizon occupied
an area of at least 160 m2. A series of AMS dates (Ammerman et al. 1992) showed that
these lowermost remains dated to between AD 550–700, thus placing the activity coeval
with the Lombard occupation of mainland Italy – essentially between the time when
Cassiodorus addressed his famous letter4 to the Lagoon dwellers in AD 537–8 and the
relocation of the ducal seat from Malamocco on the Lido to San Marco in AD 811
(Ammerman and McClennen 2001, 8 & 13). The fact that the first human use of the site
appears to have involved more than just simple wooden structures suggests a
construction of some consequence. Whatever its nature, the structure pre-dates the earliest
historically attested literary record for the area thereby demonstrating the value of
archaeological investigation for elucidating the early origins of Venice.

San Francesco del Deserto

Near Burano, in the northern Lagoon, is the small island of San Francisco del Deserto,
which has been, since the 13th century, the home of a Franciscan monastery. In 1993
148 R. A. HOUSLEY, A. J. AMMERMAN AND C. E. MCCLENNEN

archaeological fieldwork began on the north side of the church. The first excavation
uncovered the remains of a boat, which was made using ribs of oak (Quercus cf. robur L.),
a hull of lime (Tilia sp.) and held together by ‘pegs’ of dogwood or cornelian cherry
(Cornus sanguinea L. or Cornus mas L.) (Ammerman et al. 1995, 501–3). AMS dates on two
of the ‘pegs’ indicate that the boat was constructed sometime between AD 425 and AD
550. The boat came from an elevation of –1.67 m relative to mean modern sea level. At a
lower level in the same excavation, –1.80 m, a series of vertical and horizontal wooden
poles were found, which are thought to represent an enclosure or simple wharf of similar
age to the boat.
In 1998, further excavation in an adjoining area uncovered a succession of wooden
palisades, dating from the late 4th (phase II) to the early 6th centuries AD (phase X), that
are interpreted by the investigator to represent repeated attempts to reinforce and stabilise
the island bank closest to the sea (Figure 4; Ammerman and McClennen 2001, 19). These
structures provide a good example of how artificial measures were required to combat
the erosional force of the tides and waves. The three most common species of trees used
by the builders to construct these wooden palisades were alder (Alnus Miller), oak
(Quercus) and elm (Ulmus L.), the alder being mostly used for the smaller diameter upright
posts and the larger oak and elm for horizontal planking. Less numerous forms included
timber from fir (Abies Miller), apple type (Maloideae), willow (Salix L.), dogwood (Cornus),
and sea buckthorn (Hippophae rhamnoides L.). The numerous fragments of associated
broken amphorae, some of which had originated from Gaza in the eastern Mediterranean,
show how cultural material came to be re-used in Venice for secondary purposes, as
landfill or as part of a breakwater defence. While there are other archaeological finds
that place occupation on the island to as early as the 1st century AD, these 4–6th century
palisades are the earliest wetland structures so far discovered at San Francisco del
Deserto.

Torcello
The island of Torcello is where the best evidence for the early architecture in the Venetian
Lagoon has been found. The earliest structure, which is over 7 m in length, is documented
both beneath the fourth nave of the church and just on the north side of the basilica. It
consists of a walkway made of tiles, built in the 2nd century AD and resting directly on
the natural land surface of the marsh island – known locally as a barene (Ammerman
and McClennen 2001, 17). Below the portico of the church of Santa Maria Assunta are
the remains of two walls dating to the 5th century AD. By the 6th century there is good
evidence, from excavations on the north side of the basilica, for several small timber-
framed houses on stone and tile footings and with interior hearths, built along a pebble-
paved road (Figure 5; Ammerman and McClennen 2001, 17). As in the case of San
Francesco del Deserto, off-site environmental investigations have shown that the original
extent of the island was very much smaller. In addition to repeatedly reclaiming land at
the edges, over the centuries since late Roman times the inhabitants have locally raised
the land surface by over 2.5 m.
The first archaeological and literary evidence for the existence of an ecclesiastical
building on Torcello comes in the form of a stone inscription recording the arrival of the
 THAT SINKING FEELING: WETLAND INVESTIGATIONS OF THE ORIGINS OF VENICE 149

Fig. 4. This reconstruction diagram shows a series of wooden structures built to defend and
enlarge the north side of the island of San Francisco del Deserto in the period from the late 4th
century AD to the early 6th century. There are 14 chronological phases to this sequence with the
earliest waterside structure, stratigraphic unit 9054 on the left, belonging to phase II, consisting
of a row of vertical planks. The latest structure, unit 9015 on the right, from phase X, represents
a palisade made from wooden piles. The fact that the north side of the island was the one closest to
the sea could well explain why there were so many repeated attempts to consolidate the waterside
over the course of less than two centuries. (Ammerman and McClennen 2001, 19.)

bishop of Altinum and the establishment of his seat in the cathedral in AD 639.
Excavations on the west side of the basilica of Santa Maria Assunta have uncovered
remains of the first baptistery in association with window glass and coins that show it
was built at the end of the 7th century. Human burials have also been found and these
have produced 14C dates indicating interment going back to the 7th and 8th centuries
(Ammerman and McClennen 2001, 23). By the 10th century, Torcello was of sufficient
importance for the Byzantine Emperor Constantine Porphirogenitus to describe it in his
writings5 as a large emporium or trading station.
We have evidence in the form of macrobotanical remains – seeds, grain, nuts and
fruits – for the diet of the early inhabitants of Torcello (for a more comprehensive account
see Housley and Miller forthcoming). The abundance of millet (Panicum miliaceum L.)
and cereal bran (from Triticum/Secale: wheat/rye and Hordeum: barley), preserved in
150 R. A. HOUSLEY, A. J. AMMERMAN AND C. E. MCCLENNEN

Fig. 5. This reconstruction diagram shows the 6th century AD structures found during excavations
on the northern side of the basilica of Torcello. Several small timber framed houses, on stone and
tile footings and with interior hearths, are shown along a pebble-paved road. Seen at a lower level
on the left side is the 2nd century AD tile walkway that rested directly on the natural land surface
of the salt marsh (barene). (Ammerman and McClennen 2001, 17.)

midden deposits off the edge of the then smaller island, is highly suggestive of sewage
linked with the remains of past meals or discarded associated waste. In addition to bran
there are also seeds of barley and common (i.e. broomcorn) millet. Other food, or potential
food, plants from Torcello include fragments of hazel nuts (Corylus L.), walnut (Juglans
L.), stone pine (Pinus pinea L.), and possible sweet chestnut (Castanea sativa L.), the stones
from sloe (Prunus spinosa L.), and fruits from the cultivated grape (Vitis vinifera L.) and fig
(Ficus carica L.). A number of these plants can tolerate wet ground but on the whole it is
 THAT SINKING FEELING: WETLAND INVESTIGATIONS OF THE ORIGINS OF VENICE 151

likely that they were imported to the island of Torcello, albeit not necessarily from very
far away. A range of plants associated with crop weeds or as habitants of waste ground
were also present in the same samples. The presence of one seed of Vervain (Verbena
officinalis L.) could point to a possible medicinal plant for it has a long and well-
documented history as a treatment for such complaints as depression, epilepsy, and to
treat wounds and skin irritations (Stuart 1989; Podlech 1996). However, since it is native
to the Mediterranean and grows along roadsides and in waste places, there is no proof
that the early inhabitants of Torcello used Vervain in such a way.

Conclusions
The purpose of this paper has been to introduce some of the more distinctive features of
the archaeological investigations that are now under way on the origins of Venice. Thus
a start has been made to move beyond the legends that the Venetians held for many
centuries and to base our knowledge of early life in the Lagoon on these new lines of
investigation. It is clear from initial studies that even in the first phases of settlement
human manipulation of the environment was happening in small ways, and that the
cumulative effects of these interventions were to influence the subsequent development
of the city and the Lagoon. Whilst much of the historical evidence comes from a time
when Venice had acquired an urban character with its cultural, religious, political and
economic life centred firmly on San Marco, the Rialto and the Grand Canal (Ortalli
1981), the early archaeology paints a different picture of less nucleated settlement in
which power lay with a number of centres that changed with time, and where the pattern
of communication within and beyond the Lagoon was significantly different to that
prevailing in later times (Ammerman 2003). Due to the environmental setting, it is
inevitable that a significant wetland archaeological component is involved, although as
this account has shown, some of the archaeology represents dry land structures that
only subsequently became submerged through a combination of relative sea level rise
and the regional subsidence. Work has been undertaken, and is still ongoing, in a number
of fields of study that have not been discussed in this paper to any degree. For example,
the pollen and foraminifera from sediments in the Lagoon, the dendrochronology of the
timbers, the chemical composition of the early Venetian glass (of particular interest as
Venice became one of the major glass manufacturing centres of Europe), and the stable
isotope composition of the human burials, are all fields that in their own way have
contributed to our better understanding of the early inhabitants of this region. It is likely
that further excavation work conducted ahead of building restoration by the Venice
Superintendency in the next few years will provide further information on the first
Lagoon dwellers.
152 R. A. HOUSLEY, A. J. AMMERMAN AND C. E. MCCLENNEN

Acknowledgements
We wish to thank Maurizia De Min and the Superintendency of Archaeology for their
encouragement and collaboration over the years. A fair proportion of the information
used in this paper derives from a conference, exhibition and catalogue organised at
Colgate University in October 2001 (Ammerman and McClennen 2001). In addition to all
the individuals and organisations whose help enabled that event to take place, the authors
would like to acknowledge the financial assistance provided by the Arts and Humanities
Research Board (AHRB) that enabled this paper to be presented at the WARP wet-site
archaeology meeting in Olympia in 2003.

Notes
1. ‘You live in houses like those of sea birds … by weaving together flexible reeds you manage to
increase the solidity of the ground in the lagoon and you are not afraid to pit such a fragile defence
against the surging tides of the sea.’ Letter to the Maritime Tribunes
2. This is where the connection is made between the Lombard invasion of Italy in AD 568 and the
movement of refugees from mainland cities to islands in the region (Carile and Fedalto 1978)
3. So far only a few vessels have been recovered by systematic underwater archaeology in the Venetian
Lagoon. Note that in the case of the Venetian galley recently recovered at San Marco in Boccalama
the vessel (stripped of its fittings) was intentionally sunk in the 14th century in order to defend the
island from erosion.
4. Cassiodorus, Various Letters XII.24; XX.22.
5. Constantine Porphirogenitus, On the Administration of the Empire 27.71 FF; 28.

References
Agazzi, M., 1991, Platea Sancti Marci. Venezia: Comune di Venezia.
Ammerman, A.J. 2003 Venice before the Grand Canal. Memoirs of the American Academy in Rome 48:
141–158.
Ammerman, A. J. in press Flooding and environmental challenges for the Venice Lagoon. In Fletcher,
C. and Spencer, P. (eds) The Third Dimension in Venice. Cambridge: University Press.
Ammerman, A. J., De Min, M., and Housley, R. A., 1992, New evidence on the origins of Venice,
Antiquity 66: 913–916.
Ammerman, A. J., De Min, M., Housley, R. A., and McClennen, C. E., 1995, More on the origins of
Venice, Antiquity 69: 501–510.
Ammerman, A. J., and McClennen, C. E. (eds.), 2001, Venice before San Marco. Recent Studies in the
Origins of the City. Hamilton, New York: Colgate University.
Ammerman, A. J., McClennen, C. E., De Min, M., and Housley, R. A., 1999, Sea-level change and
the archaeology of early Venice, Antiquity 73: 303–312.
Brown, P. F. J., 1991, Self-definition of the Venetian Republic, in City-States in Classical Antiquity
and Medieval Italy (Eds. Molho, A., Raaflaub, K. and Emlen, J.): 511–548, Ann Arbor (MI):
University of Michigan Press.
Carile, A., and Fedalto, G., 1978, Le origini di Venezia. Bologna: Pàtron.
Crouzet-Pavan, E., 1992, ‘Sopra le acque salse.’ Espaces, pouvoir et société à Venise à la fin du Moyen
Âge. Rome: Ecole Française de Rome.
De Min, M., 1990, Venezia. Rinvenimenti medioevali nella chiesa di S. Lorenzo. Notizie preliminari,
Venezia Arti 4: 159–166.
 THAT SINKING FEELING: WETLAND INVESTIGATIONS OF THE ORIGINS OF VENICE 153

De Min, M., 2000, Venezia. Chiesa di San Lorenzo di Castello: un esempio di scavo corrlato al
restauro architettonico, in Ritrovare restaurando. Rinvenimenti e scoperte a Venezia e in Laguna: 41–
52. Venice.
Gatto, P., 1980, Il sottosuolo del Litorale Veneziano. Consiglio Nazionale della Ricerche, Instituto per
lo studio della Dinamica delle Grandi Masse. Venice: Technical Report 108: 1–19.
Housley, R.A. and Miller, J.J. in press The use of wood and other macroscopic plant remains from
the Venice Lagoon. In Ammerman, A.J. and McClennan, C.E. (eds) The Origins of Venice.
Baltimore: John Hopkins University Press.
Leciejewicz, L., Tabaczynska, E., and Tabaczynski, S., 1977, Torcello. Scavi 1961–62. Rome:
Fondazione Giorgio Cini.
McClennen, C. E., Ammerman, A. J., and Schock, S. G., 1997, Framework stratigraphy for the
lagoon of Venice, Italy: revealed in new seismic-reflection profiles and cores, Journal of Coastal
Research 13 (3): 745–759.
Ortalli, G., 1981, Il problema storico delle origini di Venezia. In Le origini di Venezia. Symposium
italo-polacco: 85–89. Venice: Marsilio.
Podlech, D., 1996, Herbs and healing plants of Britain and Europe. London: HarperCollins.
Revelle, R. R., 1990, Overview and recommendations. In Sea-level change (Ed. Revelle, R. R.).
Washington (DC): National Academy Press. Studies in Geophysics: 3–34.
Schulz, J., 1991, Urbanism in Medieval Venice, in City States in Classical Antiquity and Medieval Italy
(Eds. Molho, A., Raaflaub, K., and Emlen, J.): 419–466. Ann Arbor (MI): University of Michigan
Press.
Stuart, M., 1989, Reference section, in The encyclopaedia of herbs and herbalism (Ed. Stuart, M.): 141–
283. Novara: MacDonald & Co Ltd.
Stuiver M., and Braziunas, T. F., 1993, Modeling atmospheric 14C influences and 14C ages of marine
samples to 10,000 BC, Radiocarbon 35 (1): 137–190.
Stuiver, M., Pearson, G. W., and Braziunas, T. F., 1986, Radiocarbon age calibration of marine
samples back to 9000 cal yr BP, Radiocarbon 28 (2B): 980–1021.
Tombolani, M., 1988, Saggio stratigrafico a Torcello, in La Venetia dall’antichità all’alto medioevo:
205–214. Rome: Enciclopedia Italiana.