The Hidden Landscape of Prehistoric Greece

01 Bintliff et al.
16/11/01 3:19 pm Page 139
Journal of Mediterranean Archaeology 12.2 (1999) 139-168
The Hidden Landscape of Prehistoric Greece
John Bintliff1, Phil Howard2 and Anthony Snodgrass3

1 Faculteit der Archeologie, Universiteit Leiden, Postbus 9515, 2300 RA Leiden, The Netherlands
2 Archaeology Department, University of Durham, South Road, Durham DH1 3LE, United Kingdom
3 Faculty of Classics, University of Cambridge, Sidgwick Avenue, Cambridge CB3 9DA, United Kingdom
Abstract
For all-period intensive surveys in Greece, even those of very recent years, an abiding problem has been
the difficulty of detecting prehistoric remains, whether at the level of nucleated sites or in the form of
scatters across the landscape. The authors suggest explanations for the problems encountered in this
regard, over the past 20 years, by the Boeotia Survey. They offer some first steps towards the solution
of these problems, based on a reassessment of the actual results achieved, here and elsewhere, by inten-
sive survey methods.
Introduction sites and ‘offsite’ finds, in which the artefacts of

historic eras were dominant. The Argolid sur-
When we first embarked, in 1979, on intensive vey, begun ‘unofficially’ by Michael Jameson
survey in Boeotia, Central Greece (Figure 1), with a more traditional, topographic approach
we were joining a tradition with a decided pre- during the 1950s, but transformed into a field-
historic bias. Of the pioneer archaeological by-field, intensive survey by the early 1970s
surface surveys of Greek lands in modern times (Jameson et al. 1994) had in fact already inau-
that were our models, Messenia (McDonald gurated an approach to the Greek landscape
and Rapp 1972) was designed primarily to that did not privilege any one phase, but
reveal, in unprecedented detail, the later pre- focused instead on the patterns of ‘sequent
historic landscapes in its chosen region, while occupance’, right down to the present day. The
Melos (Cherry 1982; Wagstaff and Cherry same can be said of the Ayiofarango Survey in
1982) ran concurrently with the excavation of Crete, from the mid-1970s (Blackman and
the Late Bronze Age site at Phylakopi on the Branigan 1977).
same island. Boeotia, however, belonged to a Both in the Argolid and in the younger
younger generation of field surveys—Keos, generation of surveys that followed, certain
Nemea, Methana, Laconia—each of them recurrent features became strikingly apparent
inaugurated between 1979 and 1984. These
in the relationship between the later prehis-
were influenced by the trend in the United
States, the leading source of survey theory, toric (Neolithic, Bronze Age and Early Iron
towards ever more intensive fieldwalking and Age) and the historical (Graeco-Roman,
recording of all traces of human activity, Mediaeval and Post-Mediaeval) occupancy of
regardless of period. Newer surveys were thus what was essentially the southern, lowland
unavoidably confronted by a dense mass of Greek landscape. Prehistoric settlement ‘sites’
140
01 Bintliff et al.
16/11/01
3:19 pm
Bintliff, Howard and Snodgrass

Page 140
Figure 1. The province of Boeotia, Central Greece, showing the position of the two major areas covered by the Boeotia Survey Project (Thespiae dis-
trict in the south-centre; Hyettos district in the northeast), modern villages, and archaeological locations studied by the Project outside these
two survey zones.
01 Bintliff et al. 16/11/01 3:19 pm Page 141
The Hidden Landscape of Prehistoric Greece 141
were notably less numerous by comparison approaches were infinitely easier for the his-
with those of Graeco-Roman date, and for torical era, and it was a compensation for the
prehistory overall surface artefact densities two directors to exploit the seemingly inex-
were, for the most part, low. Demographically, haustible database available from the surface
it was difficult to believe that populations in finds of later periods, with the additional
these phases had ever been comparable with opportunities for linking them to archival and
those that could be inferred from the infinitely formal historical sources (Bintliff 1991; 1996;
more numerous remains of historic times. 1997a; 1997b; Bintliff and Snodgrass 1988b;
This aspect had been brought home very Snodgrass 1987).
clearly in a revealing analysis of prehistoric
sites from the Melos Survey by John Cherry
The Paradox: Intensive Survey and the Pre-
(1979; 1982). Here the numerous small sites
historic Settlement Pattern
scattered across the island, taken together
with the likely occupation-span of each site This was not what we had anticipated. One
and the millennia over which the total pre- aspiration of the younger generation of ‘inten-
historic site-complement had to be spread, sive survey’ project-leaders was to reveal the
made it unlikely that many of the small sites true density and complexity of the prehistoric
mapped were ever contemporary: the popula- Greek landscape. There was an explicit con-
tion at any one time, at least until the devel- trast with the picture (then widely considered
opment of the first large nucleated village at to be misleading) given by the results from
Phylakopi by the Middle Bronze Age, was the ‘extensive’ approach exemplified by the
likely to have been tiny. Minnesota Messenia survey (McDonald and
On the Boeotia survey, prehistoric ‘sites’ Rapp 1972), which had generated something
(that is, to take the generally cited definition, of a cause célèbre within the circle of Mediter-
foci of surface artefacts that were dense by ranean topographers and field survey practi-
comparison with the local background level, tioners. The University of Minnesota team
and were assumed to mark former settlements, had managed to survey a relatively vast region
cemeteries or other areas of heightened (some 3800 sq km) through the selective vis-
human activity) were distributed widely—if iting of ‘promising’ topographic features,
thinly—across the entire surveyed area (by believed from previous site discoveries to typ-
1991, some 55 sq km covered by close-order ify settlement locations during the Late
pedestrian survey at normally 15 m field- Bronze Age (Mycenaean) era. Prior inspec-
walker intervals) (for the methodology, see tion of aerial photographs, together with local
Bintliff and Snodgrass 1985). But their num- informants’ knowledge of places where antiq-
bers appeared to be so small and their distrib- uities were prominent, had speeded up the
ution so patchy (only some 35 prehistoric task of identifying places to visit. Although to
sites identified across the entire surveyed area: some extent the Minnesota field survey broke
Oliver Dickinson, pers. com.) that the infer- out of this mode of restricted visits to pre-
ences we had hoped to draw about population determined locations, the team’s analysis of
dynamics, variable uses of landscape, settle- known sites, together with the results of all its
ment networks and hierarchies were rapidly new fieldwork, seemed to reinforce the
proved unattainable, especially as many view that later prehistoric communities in
of these sites were occupied in more than Messenia lived in small-to-medium nucleated
one prehistoric period. By contrast, such settlements, largely in the same places,
142 Bintliff, Howard and Snodgrass
throughout the Bronze Age. Moreover, a walking characteristic of these surveys

favourable comparison of the catalogue of (Argolid, Ayiofarango, Boeotia, Keos, Meth-
Late Bronze Age sites with the list of place- ana, Laconia) achieved a similar goal, with a
names in the Mycenaean palace archives at site density many times greater than that pro-
the regional centre of Pylos led to the sugges- vided by the Messenian statistics. To take the
tion that the majority of such settlements had example already cited (above) of the Boeotia
indeed been identified, before and during the Project: 35 prehistoric ‘sites’ identified in
survey. The sum total of prehistoric sites— some 55 sq km gives one site for every 1.6 sq
some 300 for the region studied—was how- km, whereas the Messenian density is 1 site
ever recognized to have been reduced by site for every 13 sq km (McDonald and Rapp
burial in alluvial valleys and related processes, 1972: 15). Here, intensive survey produced
and was therefore tentatively adjusted to a an eight-fold increase in prehistoric site den-
figure closer to 400 (McDonald and Rapp sity. Overall, the discrepancy is much greater:
1972: 141). the Messenia total site density is 1 site in
The ‘New Wave’ of exponents of intensive every 12 sq km, while a typical intensive
survey in Greece now criticized, with all Greek survey finds some 4 sites in every 1 sq
the vehemence of Young Turks, the failings of km—a differential of about 50 (Cherry 1983:
the Messenia model of ‘extensive survey’. fig. 1; Bintliff and Snodgrass 1985: 136).
Sites not found in previously hypothesized But (and it is a big ‘but’) the pattern of pre-
‘favoured locations’ would escape notice; historic sites revealed was not so strikingly
small sites requiring close-order field-by-field different from that claimed by the Messenia
walking would generally be ignored; and a survey: in comparison with the Graeco-
predictable consequence of such traditional Roman settlement density, there was a rela-
methods would be to locate only the larger tively thin scatter of genuinely small (farm?)
settlements, occupied over a longer period, sites, a small number of larger hamlet-village
especially those where defensible positions on sites and, of course, the occasional semi-urban
isolated hilltops allowed rapid, ‘targeted’ vis- ‘central place’, the latter two often occupied
its as the main approach to fieldwork (Bintliff through most periods of later prehistory and
1977a; 1977b; Bintliff and Snodgrass 1985; the historical eras. Any hopes of picking up
Cherry 1983; 1984). Indeed, despite a spirited fluctuations in settlement systems, with such
defence of the Minnesota approach on the continuity of the core nucleated sites, proved
part of its practitioners (Carothers and illusory: the very possibility of demonstrating,
McDonald 1979; Hope-Simpson 1984; 1985; for example, population pressure, with such a
McDonald 1984), the post-Messenia ‘inten- poverty of prehistoric settlement size and
sive surveys’ appeared to prove their point numbers, disappeared. For certain parts of
convincingly by discovering a vastly higher lowland Greece, it could be claimed that new
density of sites everywhere else in lowland models such as that of the Argolid team
Southern Greece than the ‘extensive’ (Pope and van Andel 1984)—according to
approach had ever imagined—especially the which widespread erosion was precipitated by
small sites interpreted as farms (Classical Early Bronze Age settlement—depended on
Greek) and villas (Roman) (cf. Bintliff and the detailed map for that phase achieved
Snodgrass 1985: table 4). For later prehistory, through intensive survey. But the geomor-
the multiplication of previously known sites phology and its chronology could have stimu-
through the field-by-field, close-order field- lated the same interpretation, from the
knowledge of Early Bronze Age settlement region in general, and the immediate vicinity
already obtained through prior study. Again, of the Mycenaean palace at Pylos in particu-
one might claim that the Melos survey could lar. The latter had been, for obvious reasons,
only have postulated its ‘implosion’ model a scene of intensive research from the 1930s
(Cherry 1979; Wagstaff and Cherry 1982) for onwards: previous extensive surveys had
the decline of small rural sites across the achieved a prehistoric site density of 1 site per
Bronze Age and the rise of a single nucleated 3.3 sq km—four times the density of the
‘town’ at Phylakopi, after the intensive survey region as a whole—and the new Pylos Project
of 20% of the island. But the excavations at has raised this figure to around 1 site per 2 sq
Phylakopi, with a traditional study and revisit km. This density is not so far from those
of the many prehistoric rural sites known prior recorded by intensive survey for prehistoric
to the intensive survey and their finds (cf. sites elsewhere—one site per 1.6 sq km in
Bintliff 1977a: Part 2, Ch. 6), would have Boeotia, for example (the Pylos and Boeotia
been adequate enough to generate the model prehistoric site databases are still provisional,
proposed. but in Shelmerdine’s useful review [1997:
Very similar conclusions could be drawn 551] she lists for Early [EBA], Middle [MBA]
from the reports of one of the latest intensive and Late Bronze Age [LBA] sites, respec-
surveys from mainland Greece, the Pylos tively, the figures of 6, 11 and 14 for the Pylos
Regional Archaeological Project (Davis et al. Survey, and 19, 16 and 16 for Boeotia, over
1997; Zangger et al. 1997; Davis 1998; cf. not dissimilar areas).
Shelmerdine 1997), which is appropriately A historical line of explanation might be
set in the homeland of the original extensive sought for the fact that the density of sites of all
Messenia survey of 30 years ago. Here, too, periods in the Pylos district, even after inten-
detailed survey has failed to provide evidence sive survey, stands at little more than 1 site per
of a hitherto undiscovered multitude of small sq km, as compared with the density of about 4
sites (whether for prehistory, or indeed even sites per sq km commonly found in intensive
for the Classical age); on the contrary, the surveys elsewhere in Southern Greece. This
pattern of nucleated and long-lived sites, can largely be accounted for by the absence of
claimed by the Minnesota project for this that explosion of small farm-sites in Archaic
region, appears to be confirmed. It is true that and Classical Greek times so prominent else-
the overall site density, on the evidence of the where—although that phenomenon is clearly
preliminary reports on the Web (http://clas- detectable for the Messenian countryside at a
sics.lsa.umich.edu/PRAP.html) and in print later time, in the later Hellenistic and Roman
(Davis et al. 1997; Davis 1998), has been periods. Already, in the Minnesota Project’s
increased by a factor of 2 to 2.5: but the volume of 1972, this pattern was seen as a pre-
increase is largely in sites of historical date, dictable consequence of the prolonged Spartan
and most of the few new prehistoric sites also occupation, which held back economic devel-
have historical occupation, underlining the opment in the region; and this reading was
thesis of persistence of locations. Genuinely recently incorporated into a multiregional
small sites, of around 0.1 ha or less, remain a comparison (Bintliff 1997c) of Graeco-Roman
rarity for any period. demographic trends in the Aegean area.
Perplexing as these findings may seem for There is also an important environmental
the exponents of intensive survey, there are contrast to be established between the Pylos
special factors that apply to the Messenian area and certain other sectors of the Greek
landscape, including (emphatically) central ondly, the Pylos team also found small num-
and western Boeotia. In the former district, bers of prehistoric sherds at irregular points
the geomorphology showed unmistakable within normal transecting, not associated
signs of serious erosion of the local marls with clusters of any period. The maps so far
(Zangger et al. 1997), which must have sub- available of prehistoric ‘off-site’ finds rest
stantially affected the recovery of prehistoric merely on a limited part of the potential pre-
sites and material. Boeotia, however, exem- historic transect finds, those firmly dated
plifies a landscape that, though partly com- (Sebastian Heath, pers. com.), so that the
posed of similar marls, has remained (for final distribution will be denser, but in our
Greece) unusually stable—a fact for which we view the phenomenon of up to 20 such
have once again the authority of Eberhard findspots per prehistoric phase merits close
Zangger (pers. comm.). As, however, two of us attention. With 17 and 20 ‘off-site’ transect
pointed out in a much earlier paper (Bintliff locations for the MBA and LBA, but only two
and Snodgrass 1988a), the effect of prehis- for EBA, a strong contrast is suggested; but
toric and ancient erosion—excepting the rare the decision not to collect undecorated
cases of sites exposed on very steep slopes—is coarseware body-sherds during fieldwalking
rather to concentrate surface artefact finds as (Davis et al. 1997: 401) in our view may well
‘lagged deposits’ while preferentially remov- account for the contrast in identified find-
ing the soil fines around them. Unless other spots to the detriment of the Early Bronze
factors were active (see below), one might Age, where surface ceramics are typically
expect that Messenian surface artefacts would characterized by such sherds.
be better exposed to view than those emanat- If the latest data from the Pylos Regional
ing from the more conserved palaeosols of Archaeological Project can thus be argued to
Boeotia. fall broadly into line with the site statistics
More positively, two highly significant fea- from other intensive survey projects in south-
tures stand out from the admirably detailed ern Greece, it remains true that all such sur-
preliminary publications and Internet Edition veys have yielded a rather low representation
of the Pylos project: their importance will of prehistoric sites. Quite apart from the gen-
shortly become apparent. First, small numbers eral poverty of Neolithic settlements as one
of prehistoric sherds frequently occur within moves southwards from the tell landscapes of
collections from historical-period sites, the Thessaly, sites of Early, Middle and Late
latter being clearly recognised by fieldwalkers Bronze Age times are rarely single period; and
as sites (or ‘POSIs’ to use their term) from the they are neither sufficiently dense nor large
far more abundant historical surface occupa- enough to support any thesis of population
tion traces. Purely prehistoric sites are very even approaching local resource ceilings.
rare, compared to historic sites where a minor Here we include the major centres, which
prehistoric find component has also been bear no comparison with even a middle-rank-
recognized. In the (still incomplete) site ing city-state of Classical Greek times
gazetteer available on the Internet, definite (Bintliff 1997b). This remains true even if
prehistoric site occupation is approximately one assumes that known sites within a long
matched by findspots where the prehistoric prehistoric period were continuously occu-
pot or lithic component is so low as to be pied—which, as John Cherry cogently argued
classified as probably non-site for each or at (1979), is unlikely to have been the case for
least one recognized prehistoric period. Sec- the kind of hamlet, and rarely farm, settle-
ments that almost all these sites represent. point of first discovery, which led to a
The purpose of this article is to suggest, significant enhancement of our prehistoric
from the basis of our experience in Boeotia, assemblage, later enabled us to demonstrate
that the history of research hitherto out- the existence of a ‘site’, with EBA and MBA
lined—up to and including the latest results phases, at this spot (PP19/20). It would
from the Pylos Project—rests upon an almost certainly have eluded recognition if
entirely false picture of the prehistoric land- we had passed smoothly across it, observing
scape of southern lowland Greece. The shape only the typically high densities of historical-
of that landscape is of a totally different char- period ceramics that form a regular carpet
acter from any map of prehistoric sites so far around the large Graeco-Roman city of Thes-
produced, or deployed in favour of competing piae. Perhaps a single piece of coarseware
demographic and settlement pattern models. would have caught the eye of one fieldwalker;
Throughout this long debate over survey but, measured against the several hundred to
methods it has kept its secret—it has been a several thousand sherds that are the statistical
landscape hidden from our knowledge. average for a 60 × 50 m transect in this area
of our survey (between 74% and 85% of them
Towards a Solution of the Problem Classical-Hellenistic in date), no significance
would have been given to that observation.
The 20-year life of the Boeotia Survey has The second story has a similarly disconcert-
included a number of curious experiences. ing flavour. We had observed that two ancient
Three incidents remain firmly fixed in the (Graeco-Roman) farmsites lay less than 50-
memory, because they raised a spectre that 60 m apart, in an area of olive groves a kilo-
seemed impossible to lay to rest. First, one of metre west of the ancient city. Were they
us (J.L.B) was leading half a dozen student merely a single large rural site with surface
field-walkers along the plateau rim above the visibility problems in the intervening space?
ancient city of Thespiae (immediately south We resolved to revisit the spot: walking
of the modern village of that name in the extremely slowly and carefully back and forth
centre of Figure 1). We stopped transecting between the two previously recognized site
for a water-and-biscuits break, sitting athwart areas, we were soon able to confirm that there
our transect route on lumps of conglomerate was indeed a clear interval, lacking either
bedrock that had pierced the Tertiary marls in higher densities of ancient ceramic or freshly
this locality. As the minutes passed, we broken material. At the same time, however,
became dully aware of some curious and we again became aware of a new and hitherto
poorly-defined anomalous shapes in the unrecorded presence in this locality—pre-
ploughsoil of the field immediately before us. cisely in the ‘off-site’ interval we began to see
They were like-coloured to the soil, and the ‘wet-biscuit’ shapes of prehistoric coarse-
lacked the sharp, clear edges of normal ware. Here we had walked and found histori-
ceramic sherds. We went to investigate. As cal sites, but had recorded nothing of this
we pulled the fragments out of their soil other-worldly scatter. Once more, the pieces
matrix, their texture was moist, soft and concerned were very low in number and all
very fragile—‘wet biscuit’ is an appropriate but indistinguishable from the surrounding
description. Later, however, they hardened in soil in colour, hardness and shape; once more,
the dry summer heat and began to recover the they were numerically swamped by the abun-
shape and appearance of prehistoric coarse- dant ancient potsherds that surrounded them,
ware. Careful gridded collection around the but a prehistoric site (PP25) was in evidence
nonetheless—as later intensive collection absolute density changes were the primary
was to prove (with all three Bronze Age means of identifying sites, but with our train-
phases present). ing of students to distinguish the less common
The third story again arose from the discov- wares, in lesser concentrations, which might
ery of an ancient, Graeco-Roman farmsite signal a genuinely small prehistoric site or a
(MPA6), isolated, but rich in surface remains; historic burial location, we had still expected to
this time it was perched in a small upland correct an understandable discovery bias
plateau above the Theban Plain near On- towards the more obvious kind of rich, focused
chestos (in the north of the Thespiae Survey prehistoric and historic sites.
Area of Figure 1). When the ceramics were What emerges from the stories just related?
catalogued and dated by our ceramic experts, a Most of the prehistoric pottery in our region
tiny collection—two or three pieces out of sev- was coarseware; it seemed to survive poorly
eral hundred ancient sherds brought back as a and was hard to distinguish in the soil. The
sample—of MBA and LBA fineware was much more conspicuous wheel-made fine
identified. Such is the relative scarcity of pre- wares—Matt-painted, Minyan, Mycenaean—
historic sites in most intensive surveys that the are easy to spot in excavated material, yet the
discovery of a new exemplar is greeted with hard fact is that we found them even less fre-
some excitement. As it happened, that year quently than the hand-made ‘wet biscuit’
our prehistoric pot consultant, Chris Mee, was wares. Why was this? On the one hand, their
a Mycenaean specialist. Seized with enthusi- numerical assemblage representation is far
asm, he set out for the site, only to spend fruit- inferior to that of the coarser plainwares, and
less hours on and around it in search of a single secondly, we have very frequently observed
additional fragment. But if there was no site that once such fine wares erode out of subsur-
‘beneath’ the ancient farm, or lurking close by, face deposits into recurrently turned plough-
how could one account for the arrival of these soil, they soon lose much of their surface and
pieces within the surface debris? distinctive contours—making them hard to
At this point, it should be observed that we separate at fieldwalker height from undiagnos-
in Boeotia had formed a conventional impres- tic body-sherds. Further, since our region of
sion of what a prehistoric site ought to look Greece received very little obsidian, and the
like. Indeed, every season we had trained new local chert looked similar to the regular stone
field-walkers by taking them to our ‘model content of the soils, we lacked another obvious
behaviour’ minor prehistoric site—Onchestos/ means to compensate for the poor apparent
Kazarma on the low watershed where the showing of prehistoric ceramic clusters across
Thebes–Livadheia road passes from the The- the Boeotian land surface. To judge by the
ban Plain to the basin of the former Lake accidental discovery of such sites, after stan-
Copais. Here, although the diameter of the dard intensive fieldwalking had been unable to
‘site’ was modest (some 80 m—so not exactly distinguish them, we had to face the fact that
small), we were always sure of collecting bag- our survey was simply failing to detect their
fuls of potsherds from all Bronze Age phases— surface debris unless it formed quite dense con-
and very little else. Of course Onchestos, with centrations. But could ‘two or three sherds
its ease of access, its striking topography, and gathered together’ constitute a case for the
that very quality of inexhaustible prehistoric existence of a vestigial prehistoric site?
surface material, had been spotted long before The explanation based on the increasing
our survey began. In normal fieldwalking, invisibility of prehistoric surface sites with age,
even to intensive survey approaches, was a in a landscape dominated by far denser clusters
hypothesis that we had put forward tentatively and off-site spreads of historical ceramics.
in preliminary publications on the Boeotia sur- A good example of this type of site is our ‘Val-
vey (Bintliff and Snodgrass 1985: 137-38; ley of the Muses 2’ location (this district lies in
Bintliff 1985: 214-15). We emphasize surface the western part of the Thespiae Survey Area
ceramic scatters, because it is precisely in the on Figure 1, directly west of Palaeopanagia vil-
effect of surface exposure to mechanical and lage). Figure 2 (upper) shows the density of sur-
chemical breakdown, through cultivation and face ceramics recorded on the site grid after
the weather, that ploughsoil assemblages stand visibility correction (i.e. multiplying up the raw
to diminish rapidly in comparison to those number of sherds seen and counted by the esti-
revealed in excavated levels. This concept drew mated percentage of bare soil visible), as well as
support from suspicions voiced by another pre- (lower left) the complete dominance of closely-
historic Aegean ceramicist (Rutter 1983), con- datable sherds from the Archaic–Early Hel-
cerning the poor ‘visibility properties’ of certain lenistic period in the small sample (138) of
prehistoric periods, leading as a consequence to ceramics taken from the site. The clearly
an inadequate record of sites from those peri- focused distribution for the latter phase points
ods. At the same time, we became aware of a to the likely position of a farmhouse structure
close parallel between observations made from on the east side of the grid. In scale and surface
surface survey in Italy, and one of the charac- ceramic density, this is characteristic of the
teristic phenomena of our site collections. In ubiquitous type of Classical rural site argued to
Boeotia we not infrequently found that a well- represent a ‘family farm’. Also in Figure 2
defined ancient or mediaeval site, like that (lower right), however, we see that, associated
mentioned in the third story above, revealed, with the dominant occupation phase, there is a
after gridding and sampling, a small number of small cluster of LBA pottery, only recognized
prehistoric potsherds in among its plentiful his- during the laboratory study of the sample col-
toric assemblage. How was one to interpret this lection. Beyond doubt, this was also a local
phenomenon? Di Gennaro and Stoddart focus of Mycenaean activity; the scant finds in
(1982) reported the same from their re-exami- the sample collected (10 Mycenaean sherds
nation of the vast find-collection from the area against a total of 53 for the main historical
of the South Etruria Survey, north of Rome. As period) probably reflect the impoverished frac-
in Boeotia, targeted revisiting and re-study of tion remaining in the ploughsoil from a farm-
the collections of a number of historic sites, house of similar scale and population to the
each with a seemingly insignificant additional later historic ‘family-scale’ farm site. Had we
component of putative prehistoric material, not identifed an ancient farm in the same loca-
had revealed a number of unsuspected prehis- tion, it is statistically very likely that the earlier
toric sites; closer inspection of the finds and the cluster would have entirely escaped notice
sites elevated the small prehistoric component when the fieldwalkers passed over the locality.
still further, although total recognized numbers If an interpretation along these lines is
were far inferior to those of the historical mate- accepted, then our site maps of prehistoric
rial. Back in Greece, in a sophisticated treat- Greece may well be so far from past reality as
ment of the Keos Survey surface assemblages, to undermine the many models based on
Cherry et al. (1991: 222-23, fig. 9.7) have also them—in particular, those concerned with
illustrated statistically the way in which prehis- the distribution and scale of population.
toric surface finds suffer adverse discrimination, Merely to recognize the possible existence of
the ‘hidden landscape’ of our title, however, Mediaeval village of Askra in the Valley of the
may reveal a failing on the part of recent Muses is especially illuminating as a contrast to
intensive surveys in Greece for the prehistoric such almost invisible prehistoric sites as
era (our own included), without providing a PP19/20, PP25, MPA6 and VM2, though it is
way forward in dealing with the problem. the latter which may nevertheless represent
The case of the large Graeco-Roman and the normal pattern in southern Greece. At
Figure 2. Boeotia Survey, Valley of the Muses site 2. (Above) total surface pottery density (visibility-corrected count
across the site grid). (Below) distribution of finds from a small sample collection made for dating (n = 138):
lower left, Archaic–Hellenistic (a-ehl) finds, the dominant phase; lower right, LBA (lh) finds.
Askra, in the Classical period and then again small numbers of prehistoric sherds
in Late Roman times, datable ceramics indi- among the pottery samples from highly
cate a maximal expansion of the site, which at visible historic sites.
some 11 ha could have contained over 1000 3. Among the causes for this situation are
people (Bintliff 1997b). Yet the distribution of the poorer survival potential, in plough-
EBA ceramics is also quite extensive, as our soils as continuously cultivated as those
on-going analysis has shown. Apart from small of Boeotia, of prehistoric ceramic—most
quantities found over the whole site (under 3% of it coarseware; the much longer period
of the total site sample collected for all peri- over which taphonomic processes have
ods), the northwest sector has a nucleus of operated to destroy the greater part of it,
some 2-3 ha, where prehistoric sherds comprise and reduce the remainder to poorly rec-
over 40% of the sample collected in that area. ognizable, low density patches across the
There was no risk whatever of such a concen- plough-soil surface; finally, and perhaps
tration being missed. Here then is a traditional especially in Boeotia, the swamping of
‘prehistoric settlement’, whose size and artefac- the landscape by very high densities of
tual density divert attention from the far more historical sherds, making it almost
widespread, minimal-density prehistoric sites impossible to detect or note as poten-
that may lie all around it in the Valley of the tially diagnostic these small elements
Muses, almost all of which we now believe to through standard fieldwalking, especially
have eluded our intensive field-by-field survey when they have been reduced to barely
of the district—unless, like VM2, they happen visible or invisible body-sherds.
to lie beneath the more visible scatters of his- The chance revelation in the field of what
torical sites. are probably the more typical, but vestigial,
sites, is too rare an event to constitute a
Windows on the Hidden Prehistoric Land- viable means of evaluating the scale of ‘hid-
scape of Southern Greece den’ prehistoric settlement. An initial step
towards producing a more realistic prehistoric
Three main propositions emerge from the
map can be achieved through a retrospective
argument so far, based on our experience of
upgrading of the seemingly insignificant col-
survey in Boeotia:
lections of prehistoric sherds from later,
1. Generally, only the larger, multi-phase higher-density sites. Rather than taking these
prehistoric sites achieve immediate recog- tiny scatters at face value—which is what led
nition, through both extensive and inten- Jacobsen (1984: 33), for example, to suggest
sive survey, and hence constitute the that the small clusters of Neolithic activity
present ‘prehistoric site database’ and the traces at the ancient sanctuary of the Argive
model for what such a site should look like Heraeum testified to seasonal pastoralism
in the field. rather than permanent settlement—we would
2. Secondly, by unintentional and indirect wish to argue that such vestigial surface traces
means, one may come to suspect the are all that we would expect from a typical
presence of a hidden multitude of small prehistoric settlement, particularly one
smaller, less continuously occupied prone to heavy weathering and massive later
sites—either through the kind of reoccupation. But here again, this is merely to
serendipitous discovery related above, or detect a symptom of a potentially far larger
from the repeated experience of finding phenomenon, biased by the coincidence of
prehistoric settlements or other activity foci together with surface visibility, but also col-
with historic settlement sites. We wish to sug- lected a sample of the datable surface material
gest a more powerful means, one less depen- from ‘off-site’ areas. The field-walkers, survey-
dent on sheer chance or the association with ing a 2 m wide strip at 15 m intervals between
later occupation-sites, of recapturing the geo- walkers, and recording a count of every visible
graphic scale and frequency of prehistoric artefact, were also instructed to collect a small
occupation in lowland Greece. sample of such sherds as appeared especially
Figure 3 displays one of two districts within diagnostic for dating (feature sherds, varied fab-
the area surveyed by the Boeotia Project, where ric and decoration types). The district shown,
from 1989 onwards the field teams not only car- the Leondari-Southeast and Thespiae-South
ried out our standard procedure of recording transect block, comprises some 5.2 sq km
surface sherd density for every transect walked, immediately abutting the walls of the ancient
Figure 3. Fieldwalking transects in the LSE/THS sectors, adjacent to the ancient city of Thespiae, an area of 5.2
sq km. The location of the 18 rural sites identified during total fieldwalking of these sectors is indicated,
with the area of the survey grid over the city in grey.
city of Thespiae (the city lies north of the tran- dismissed as a ‘non-site’; all the rest are Graeco-
sects and its southern perimeter is enclosed by Roman, one having Mediaeval settlement evi-
them on three sides, and is itself just south of dence too; but none had more than a small
the modern village of Thespiae marked on Fig- scatter of prehistoric lithics or ceramics, and we
ure 1). Displayed on the transect map are the thus have no official ‘prehistoric sites’ in this
location of the 18 small rural sites identified area.
during the fieldwalking of this district. From Figure 4 maps 46 prehistoric lithic finds from
subsequent analysis, one of these, at first the sample of surface artefacts brought back by
believed to be a potential historic site, was later fieldwalkers from the transects of this district
Figure 4. Fieldwalking transects in the LSE/THS sectors, adjacent to the ancient city of Thespiae, with transects
containing 1-2 lithics from the dated sample collection of offsite finds shaded. Complementary finds of
prehistoric lithics from the on-site sample collections at historical sites are shown as numbers of finds
attached to each site location dot.
(the relevant transects have shading keyed for by the Kanavaris stream in the northeast.
finds density), together with another 46 lithic Eleven out of the 18 putative historical sites,
finds recovered during gridded collection on however, also yielded prehistoric lithics. As
the 18 historical sites (shown as numbers of noted earlier, obsidian is rarely utilized in our
finds in boxes attached to site-location dots). region, in contrast to local chert, while the lat-
No clear clustering is visible within the off-site ter is almost impossible to observe when similar
lithic finds, apart from a slightly denser group unworked stone litters the surface; fieldwalkers,
Figure 5. Fieldwalking transects in the LSE/THS sectors, adjacent to the ancient city of Thespiae. Reconstructed
total density of surface finds, visibility-corrected (total 1.37 million potsherds and lithics). Key: density
of surface finds per ha; identified rural sites shown as white circles. The sampling grid for the adjacent
ancient city of Thespiae to the north is indicated by cross-hatching.
too, are in general habituated to focusing on is done, the total recorded density count of
ceramics, to the exclusion of soil and its stone 95,895 artefacts rises by extrapolation to a
content. But in the two seasons in question total of between 1 and 1.5 million off-site sur-
(1989 and 1991), the vast majority of the face pieces within this transect bloc as a whole,
lithics were found by one single fieldwalker, omitting the gridded site-collections (Figure
who had a trained eye for prehistoric lithics, but 5). So, the small sample of some 3714 artefacts
who could only cover one-fifth of the area that the field-teams actually collected at the
walked. It is thus a safe inference that the num- same time as recording the density can be used
ber actually picked up was only a fraction of to estimate a plausible scale of total distribu-
what might have been collected; and the tion by finds-category: the multiplier should be
recorded distribution is in turn probably a much of the order of 270-400. Of the 3714 ceramic
smaller fraction, as well as an arbitrary one, of and lithic pieces collected ‘off-site’, 46 were
all the worked prehistoric lithics that can be prehistoric lithics. On the basis of proportional
assumed to lie on the surface. Since the entire recovery (lithics comprising a little over 1% of
artefact sample of ceramics and lithics, of all surface artefacts), we could postulate that
periods, collected according to our new proce- within this 5.2 sq km district there ought to be
dure from all the off-site transects in this dis- something between 12,500 and 18,500 lithic
trict, amounts to only a few thousand artefacts, pieces that might have been brought back to
when many an individual transect can boast a base, had all the ‘off-site’ surface artefacts that
figure of the same order in its ceramic density were counted been collected: that is, between
count, the likely density of any type of artefact 24 and 35 pieces per ha. This density is com-
in this area will be immeasurably higher than parable to prehistoric lithic densities of later
the quantities shown. The larger numbers of prehistoric age from surveys in Southern Eng-
prehistoric lithics recovered on the surface of land (Schofield [1987] cites lithic densities of
the historical sites result from the greater inten- 8-20 per ha, and locally higher). But in reality
sity of surface inspection applied by field teams the visibility of lithics is low in a ceramic-
to the small grid units set up for sample collec- focused survey, and we should envisage an
tion across sites. Yet calculations of the additional multiplier for the suspected real
increased intensity of on-site as against off-site density of surface lithics: even if fieldwalkers
recording, in respect of the same transects, indi- had collected all the artefacts that they
cate that a multiplier of no more than 2-3 counted, the recovery of lithic pieces would
should be applied. This suggests that much still be lower than the surface reality.
higher lithic counts across the entire surveyed In the district around ancient Hyettos city,
district are likely in reality. the other area where we similarly experimented
The following numerical considerations will (from 1989 onwards) with an off-site sample
now be applied. First, the walkers conducting collection strategy for dating purposes, as well
the off-site survey directly observed only as recording total artefact density, standard
13.3% of the surface in their numerical count- intensive fieldwalking produced a similar den-
ing of surface artefacts. Secondly, adjustment sity of lithic pieces collected. Here, however,
needs to be made for variations in visibility, we additionally deployed a lithic specialist to
based on vegetation cover, in each transect (cf. walk a 2 m strip behind the fieldwalking team
also Whitelaw 1991, for Keos). This means in a sample sector, ignoring all ceramics, and
raising observed counts by a larger multiplier found a density that would be extrapolated to
for each decrease in visibility level. When this 50 artefacts per ha, giving us a multiplier of 1.5
between standard fieldwalking and expert away from the Mediterranean area. Thus,
search for lithics, at the same speed. This would research in Bohemia suggests a lower ratio of,
imply a figure of between 19,000 and 27,000 typically, 5-15% surface finds to ploughsoil total
lithic pieces for our 5.2 sq km block, to allow for (Kuna et al. 1993: 118; Kuna, in press). The
the enhancement observed at Hyettos. As total ceramic content of the ploughsoil to this
noted earlier, the arbitrary ‘windows’ of closer depth in our selected district would therefore,
surface study, opened up through intensive col- on the first calculation, rise to some 6-9 million
lection across the historical sites of this district, potsherds, and the lithic content—using the
provide further confirmation of a near-ubiqui- figure based on dedicated lithic fieldwalker
tous lithic carpet, as well as demonstrating how recovery—to some 112,000 to 160,000 arte-
lithic recovery improves through increased facts (or an average of 20-25 discards per year,
attentiveness. six per person for a single nuclear family culti-
A density of this order of lithic pieces need vating the area during farming prehistory).
not represent an unusually high level of discard. These figures, which would be even higher if
The (almost) indestructable nature of lithics the alternative estimates for the ratio of surface
would encourage very high cumulative totals, to subsurface were preferred, are not implausi-
even for tools discarded on to the original land- ble activity-rates for an area of prime fertility
surface. One cannot, in deference to earlier (excellent soils, perennial water supplies, easy
research (Runnels 1982), exclude the possibil- communications), such as the sector under
ity of continued use of stone tools into histori- study.
cal times. But it does seem likely that this All of this underlines the surprising signi-
particular spread of lithics, invariably in the ficance of the 50 or so lithic artefacts picked up
form of blades and small scrapers characteristic by standard fieldwalking, implying a sustained
of the Neolithic and Bronze Ages, and usually and intense level of prehistoric activity in this
of Melian obsidian, was mainly generated dur- district. The extrapolated surface density would
ing prehistoric times by a small, initially be very much like that predicted from targeted
hunter-gatherer and then predominantly surface lithic surveys in, for example, prehistoric
mixed-farming population in this area, between lowland Britain, which has been described by
say 8000 and 1500 BC. This would amount to John Schofield as virtually a continuous lithic
up to four lithic artefacts a year; and even if carpet (Clark and Schofield 1991: 104). Indeed,
only one family normally occupied this district, where intensive lithic survey has been applied
that would represent one lithic piece discarded to a district of very focused prehistoric activity,
per person per year. But there is a further factor as for example around Stonehenge (Richards
to be incorporated. It has been calculated 1990: 11), even the recorded lithic density may
(Reynolds 1982) that in long-cultivated agri- rise to well over 1000 pieces per ha. In Greece,
cultural soils, especially those with a relatively too, maps generated by intensive survey (e.g.
stable geomorphology like Boeotia, something Cherry et al. 1991: 40-45, fig. 3.4) have shown a
on the order of 16% of the ploughsoil assem- similar quality of continuity of spread, if natu-
blage is likely to be present on the surface at rally not (for the reasons just outlined) such
any one time, implying a likely multiplier, for high densities.
the ploughsoil population to a depth of 20 cm, What would such a vastly denser lithic distri-
of around six. For comparable investigations of bution represent in terms of prehistoric activity?
the relationship between surface and sub-sur- Two things are well established in lithic studies:
face densities of artefacts, we have to move first, in contrast to prehistoric ceramics, lithics,
while not wholly indestructible, are good sur- number of windows of information onto our
vivors in the ploughsoil. Secondly, prehistoric ‘hidden prehistoric landscape’ than are pro-
farmers, herders and hunters made and deposited vided by those chance discoveries, and retro-
lithics in a wide variety of situations—on perspective revaluations of historical sites,
manent and temporary/seasonal domestic sites presented earlier. In the Czech Republic, as in
and other activity foci, in burials, and during Boeotia and in fact much of Europe, prehistoric
extramural work. In a fertile landscape subject ceramic assemblages are dominated by coarse-
to some six-and-a-half millennia of activity by wares, with a very low survival and recognition
hunter-gatherer and mixed-farming communi- rate in the ploughsoil. For comparable distribu-
ties, such as lowland Boeotia, we would expect tions of prehistoric pottery produced by inten-
to find an almost continuous carpet of discarded sive survey to that shown here, the Prague team
lithic artefacts. Parallel research from intensive argue that only one of the two classes of tapho-
survey in Central Europe by the Institute of nomic environment is likely to be responsible
Archaeology in Prague (Kuna 1991; in press; for virtually all the current surface ceramic
Kuna et al. 1993; Salac 1995) has led to the finds. Prehistoric pottery deposited on the
insight that such widespread lithic scatters will ancient surface or in the ploughsoil (e.g. by
emanate from two classes of taphonomic envi- non-residential land-use discard, or through
ronments—pits, middens and burial structures manuring with domestic refuse) would by now,
or subsurface features currently being ‘tapped’ by almost without exception, have been destroyed
ploughing; and palaeosols where artefacts were through cultivation and natural weathering on
dropped or left on or near the surface of the soil the surface. What survives in today’s ploughsoil
and are now incorporated into the contempo- has to be almost entirely derived, and that rela-
rary ploughsoil. What we are seeing, therefore, tively recently, from a subsurface feature, a
in Figure 4, is a minuscule fragment of a land- reservoir where it has avoided the more
scape of lithic deposition, combining finds from extreme destructive forces through long burial
both prehistoric domestic and burial sites, and within an archaeological deposit—such as a pit,
‘offsite’ economic activities. a ditch, a buried midden or a tomb. (For a
In contrast, the data shown in Figure 6 indi- closely comparable view, see the evidence from
cate the distribution of the 52 prehistoric the Stonehenge Environs survey [Richards
ceramic finds collected within the off-site tran- 1990: 25-30, 228-29, 280], and that from the
sect sample of 3714 surface artefacts (the rele- A41 Project, also in England [McDonald 1993],
vant transects shaded by keyed numbers of and several other survey and excavation pro-
finds), and the 25 prehistoric sherds collected jects in England such as Maxey, Shiptonthorpe
during study of the historical sites (numbers of and Baldock [Jeremy Taylor, pers.com.] In
finds in boxes attached to site location dots). Greece, Whitelaw [1991], in his analysis of the
Here a different line of explanation has to be prehistoric surface material from Kephala and
followed. According to the models of our Paoura on Keos, has likewise argued for system-
Czech colleagues, importantly linking surface atic degradation of prehistoric surface ceramics
study to subsurface excavations at the same as opposed to lithics, suggesting that surface
locations, these ceramics should emanate from potsherds are therefore a more reliable indica-
a much narrower taphonomic context (Kuna tor of underlying prehistoric deposits, as they
1991; 1998; in press; Kuna et al. 1993; would not survive significant displacement.
Neustupny 1998; Salac 1995). Herein, indeed, Essentially, therefore, the distribution of Figure
lies the possibility of opening up a much greater 6 should be that of prehistoric sub-surface struc-
tures or sites. This quite revolutionary idea is at As we have seen, the low numbers are in
first sight counter-intuitive, as it has to be set fact entirely to be expected, first as a result of
against the extremely low numbers of pieces the low survival rate of prehistoric ceramics
recorded and their almost unpatterned distribu- compared to those of historical age in the
tion across the landscape. same ploughsoil assemblage, and secondly
Figure 6. Fieldwalking transects in the LSE/THS sectors, adjacent to the ancient city of Thespiae. The location of
all prehistoric ceramic finds identified within the sample collection from off-site fieldwalking is shown by
transect (shaded by counts per transect, see key). Complementary on-site finds of prehistoric ceramics from
the sample collections at historical sites are shown as numbers of finds attached to site location dots.
because of the very low percentage of all sur- trained in advance in the appearance of pre-
face ceramics collected for dating, compared historic, Graeco-Roman and Mediaeval to
with the absolute numbers to be found on the Modern potsherds, and in how to collect on-
surface. Even the subsurface reservoirs of site and in special off-site collections such as
ceramics of prehistoric age have undergone this one, samples of feature pot (rims, bases,
massive destruction from weathering in pre- handles), as well as a range of decorated wares
historic and later times, and we can suggest and fabric types for body-sherds. Each team
that in long-cultivated landscapes these sub- had an experienced leader providing quality
surface features may be poor remnants of control and advice on surface finds as they
much richer and deeper pits and occupation appeared. We will argue later in this paper
features largely ploughed out long ago, or that the statistical properties exhibited by
even secondary, residual deposits in structural prehistoric sherds and lithics in survey con-
artefact reservoirs of later periods (Kuna, in texts other than ‘traditional’ denser findspots
press; Salac 1995). The same Czech research recognizably forming ‘prehistoric sites’, recur
shows that long-lived sites, or those with outside the Boeotia Project wherever simi-
significant reoccupation, will serve as zones larly detailed and intensive recording has
for the enhanced preservation of prehistoric taken place, so that we feel justified in
pottery, through the physical sealing of earlier hypothesizing that the explanations for these
layers, or the reincorporation of older mater- phenomena are not confined to Boeotia, but
ial into later structural features. Short-lived, may well be of much wider relevance in low-
single-phase sites with a dominance of coarser land Greece. Likewise, although prehistoric
ceramics suffer far more severe degradation of coarseware may vary in other survey projects
their assemblages through prolonged cultiva- from being comparable to the ploughsoil
tion and weathering of the land-surface over- porous softness of Boeotia (‘wet-biscuit’) to
lying them, so widening the apparent contrast gritty toast (e.g. in Methana—Chris Mee,
for the field surveyor confronted by the intu- pers. com.), we would argue that the destruc-
itively ‘site-like’ surface appearance of larger tive forces of the weather and millennia of
and/or multi-period settlement sites, in com- cultivation bring all such material into mini-
parison to the seemingly insignificant scatter mal surface quantities alike to each other,
of a handful of prehistoric sherds representing when it emanates from a short-lived farm or
a vestigial, short-lived farm-site or similar hamlet rather than from the much richer and
activity focus. far better conserved subsoil reservoirs to be
As was earlier demonstrated, justifiable associated with long-lived village settlements.
extrapolation, even from the low numbers of Once more, what would this distribution map
prehistoric finds brought back in the selective for prehistoric finds—this time ceramic—look
sample collection throughout the Leondari like if we were to attempt an extrapolation
South East and Thespiae South [LSE/THS] process? Working on the basis of proportional
zone, allows us to infer a considerable low- representation, we can suggest that were the
density carpet across the entire survey sector. entire 5.2 sq km to have been walked at 2 m
To what extent, however, is the low sample intervals, and all counted finds collected, allow-
percentage of prehistoric as opposed to his- ing for a visibility correction, Figure 6 would be
toric finds affected by the attitude and skills of displaying for the offsite transects alone some-
the fieldwalkers, and in particular those on thing of the order of 19,000 prehistoric pot-
the Boeotia Project? The field teams were sherds at a minimum (out of our extrapolated
total surface ceramic of 1 to 1.5 million pieces). sive kind across gridded site surfaces.
For the full depth of the ploughsoil, again, a The ‘windows’ opened up through gridded
multiplier of six or more to this figure could be collection on the historical sites again pro-
postulated, to give a three-dimensional view to vide confirmation of the ubiquitous presence
this distribution. Since we have argued that sur- of prehistoric ceramic, and hence of vestigial
face prehistoric ceramics are often ‘invisible’ to sites, throughout the district. Rather than
fieldwalkers in areas of dense historical-age pot- seeing the locations of historic sites as ‘spe-
tery, or at least reduced to worn body-sherds of cial’ also for prehistoric small-site locations,
apparent undiagnosticity—unless (as at Askra) we are arguing that the recognition of low-
they too are found in dense concentrations— density prehistoric artefact scatters is purely a
this again must be a minimum, which would mechanical result of the greatly enhanced
require a further significant multiplier, to envis- attention given to the less conspicuous sur-
age the actual density of the surviving surface face finds during on-site collection in these
prehistoric ceramics, were we able to see them places. The ‘reconstructed’ pattern of prehis-
during survey (there is on average one sherd for toric sherds would probably cover much of
every four sq m of landsurface in this district, the landscape, but everywhere in small quan-
but 97% are recorded in sample collections as tities. Again, we believe it appropriate to bor-
of historical age). For this reason, whenever row from our Czech colleagues, this time their
John Bintliff and Oliver Dickinson revisited interpretative model for this striking land-
historical sites to enlarge the collection of a scape phenomenon. They have argued that
very sparse prehistoric component, the chief such a widespread but thin carpet of prehis-
method was to lie full-length on the ground toric ceramics arises as the result of a process
and inspect small areas by eye along the plane of continuous horizontal displacement of
of the soil surface, seeking for the tell-tale ‘wet small farm and hamlet sites around fertile dis-
biscuit’ shapes and textures. A similar range of tricts of the landscape over long periods of
estimates to those we have made for missed time. The likely explanation for recurrent
lithics would be applicable to ‘invisible’ prehis- relocation of small prehistoric rural sites (in
toric coarseware. clear contrast to the far rarer, larger, longer-
It will be observed that no strong correla- lived sites that comprise the traditional ‘pre-
tion yet appears between the admittedly tiny historic site’), is a long-fallow system, where
sample of lithics and that of prehistoric farmers shifted the location of their farms to
ceramics found and collected in offsite tran- rest soils and take advantage of less intensely
sects (Figures 4 and 6). Whether this could cultivated areas in the vicinity, perhaps every
have been predicted on theoretical grounds— few generations or even more frequently.
in that the lithic scatter should reflect both House structures might last one to two gener-
on-site and off-site activity, but the pottery ations, and arguably micro-locational shifts to
on-site activity only—it is noticeable that closely adjacent field plots around a landscape
better correlation occurs in the (again) tiny of broadly similar agricultural potential would
sample of prehistoric finds from the collec- ultimately lead to an almost continuous dis-
tions on historical sites in the district: nine of tribution of surface artefacts, whose ceramic
the 18 historical sites sampled provided both traces were rapidly reduced by later cultiva-
prehistoric ceramics and lithics This must in tion and weathering.
part reflect the closer attention of fieldwalkers We must emphasize the considerable risks of
during collecting processes of a more inten- chronological generalization based on sub-
samples of the very tiny samples of prehistoric millennium BC. The latter are marked by the
material collected on- and off-site in the creation of hierarchical formal field systems
LSE/THS sector. Yet a consistent pattern and are interpreted as representing a change
might be hypothesized on the data available, to short-fallow, more intensive land use and a
for further testing on a much larger sample. In new emphasis on property rights. Closer to
contrast to the multi-period occupations char- Greece, the Neothermal Dalmatia Survey
acteristic for the ‘prehistoric sites’ found by Project (Chapman et al. 1996), in present-day
both extensive and intensive survey on our Croatia, has evidenced a very dramatic reduc-
own and other regional surveys in southern tion in identified small rural sites and ‘off-site’
Greece, the prehistoric ceramics amenable to transect finds between the Bronze Age and the
dating by phase from the rural sector in this Iron Age, in parallel with a marked transfor-
present investigation, as so far studied, show mation towards a more hierarchical political
the following statistics. The finds from histori- and settlement system in the region. Chap-
cal site collections provided EBA from nine man and his colleagues (1996: 259-64) see the
sites, MBA to LBA from three; the finds from ‘off-site’ Bronze Age finds as indicative of
off-site transects datable by phase were only of either non-residential discard in a heavily
the EBA and LBA, the former predominating utilized countryside, or the debris of a short-
by about 3:1 over the latter. Very tentatively, term shifting settlement system much along
we might suggest that small-site, shifting agri- the lines of the model favoured in this paper
culture is especially important in the Final (indeed, the typical surface ceramics of later
Neolithic-EBA. In the MBA and LBA, there is prehistoric Croatia, with their characteristic
far more emphasis on larger settlement sites of coarse fabrics, would in our view make the lat-
hamlet-village size (none from this sector—but ter interpretation easily the most likely). The
several ring our study district here—see Figure deeper social and ideological meanings drawn
7), alongside a less widespread small farm-site from this process of transformation in the
network in the countryside intervening landscape may therefore, in our opinion, offer
between such minor foci. important future insights into the formation of
Other surveys (e.g. Melos, Argolid, Meth- later Bronze Age hierarchical societies in low-
ana) have produced similar on-site and off-site land southern Greece (although, of course, it
evidence for greater nucleation of settlement is not necessary to assume that such an
and a reduction in dispersed minor rural site implied pattern holds true everywhere in this
activity between the EBA and MBA-LBA, region).
which may become greatly enhanced if the Two final observations might be made. First,
implications of our reinterpretation of small it is quite possible that similar factors could be
prehistoric scatters were generalized. In south- at work in later phases of landscape occupancy
ern England, Barrett (1994) has identified a in southern Greece, if coarseware could be
similar long-term trend, contrasting a domi- shown to have formed a dominant component
nance of long-fallow shifting settlement sys- in the domestic assemblage—for example, the
tems leaving minimal artefact debris during ‘Dark Age’ of Sub-Mycenaean to earlier Geo-
the Neolithic and Early Bronze Age (fifth to metric times, or the ‘Dark Age’ of the Early
third millennia BC) (rarer larger and long- Byzantine centuries; if so, considerable revi-
lived foci excepted), with more stable nucle- sion would be required of present views on
ated settlement systems commencing in the the location and density of population foci.
Middle to Later Bronze Ages of the second Secondly, we have excluded from our range of
Figure 7. Distribution of prehistoric ceramic finds from sample collections in the on-site (numbers per site in
boxes) and off-site (shaded transects) areas of the LSE/THS sector, in relation to recognized prehistoric
‘sites’ in the surrounding district. The ancient city of Thespiae immediately adjoining this rural survey
sector to the north has rich multi-period prehistoric material from the Magoula hillock (neo = Neolithic,
eh = EBA, mh = MBA, lh = LBA) and also from the wider area of the city west of the Magoula. The
Bronze Age settlement on the hill of Palaeokarandas is several hundred metres southeast of the outer
limit of the rural survey sector and has finds of all three bronze age periods. The large prehistoric site at
Askra is approximately 5 km west of ancient Thespiae city, and has Neolithic to LBA finds.
analysis the possibility—indeed, likelihood— findspots gave less than five diagnostic sherds
that in certain periods of Greek prehistory and (with five having only one recorded sherd).
history the levels of ceramic production and Mee and Forbes (1997: 42) suggest that the
consumption varied independently of simple latter 26 findspots have insufficient finds to
population numbers (Millett 1991). denote occupational status, especially given
the strong local diagnosticity of EH coarse-
ware, and should therefore be attributed to
The Hidden Landscape Uncovered: Impli-
‘off-site’ activity or ‘scatter’. Nonetheless, they
cations
note the discordance between the location of
If we were to accept the entirely different pic- some of these ‘field’ scatters and that of
ture of the surface traces of prehistoric settle- accepted ‘sites’. Following the model proposed
ment and related activity foci which has been in the current paper, we would suggest that
put forward, it would provide us with a com- many if not most of the 26 EH ‘scatters’ actu-
pletely new way of making sense of intensive ally represent vestigial occupation sites of
survey data. We have suggested that, at the shifting settlement, small-site character. In
very least, it will be worth testing in the rest of the later Bronze Age periods, the following
lowland Greece (the Pylos district included), statistics are noteworthy (Mee and Forbes
the insights achieved over two decades in 1997: 51-53): for the MBA (Middle Helladic
Boeotia, and over an equivalent period by our [MH]), there were three findspots with more
Czech colleagues, into the taphonomy of pre- than five sherds, while one findspot had less
historic surface sites. We shall therefore con- than five sherds; all four had historic period
clude this study by introducing relevant data finds on them. Nonetheless, the authors doubt
from other recent intensive surface surveys in the occupation status of all but one of the
southern Greece, where a similar revisionary locations with more than five MH sherds, due
interpretation of prehistoric settlement activ- to low find numbers. For the LBA (Late Hel-
ity might tentatively be hypothesized. ladic [LH]) all eight findspots with LH sherds
In a forthcoming study of Neolithic site evi- were also historic sites; there were five
dence from Peloponnesian field surveys, findspots with more than five LH sherds, while
Cavanagh (1999) reports that the Laconia there were two further locations with a single
Survey has identified some 12 locations with LH sherd recovered.
Final Neolithic-Early Bronze I evidence. Of We would be very much more positive con-
these, just one findspot has significant artefact cerning the site status of all the MH ceramic
numbers, the rest consisting of very few sherds findspots, and underline the importance of the
or merely lithic finds. Cavanagh suggests that easier recognition of historic surface sites in
the latter group could either represent loca- leading to the recovery of associated prehistoric
tions without permanent occupation, or— occupation in both MH and LH. We suspect
significantly—those where severe destruction that a policy of continuous surface collection in
of prehistoric ceramics has occurred. transects would have yielded a far greater num-
On the Methana Survey, the authors report ber of small scatters from all Bronze Age eras, as
51 findspots with EBA (Early Helladic [EH]) in Boeotia, but note the significant trend in the
finds (Figure 8, from Mee and Forbes 1997: fig. findspots recorded for settlement nucleation
4.1). Twenty-five of these delivered more than after the Early Bronze Age.
five sherds of that period, and are considered For the Keos Survey, the authors map an
to be full sites; almost half of these were also impressive number of findspots of lithics and
historic sites. The remaining 26 EH ceramic prehistoric ceramic (Figure 9, from Cherry et
Figure 8. Methana Survey: findspots with EBA ceramics, distinguished by find quantity and overall site size (Mee
and Forbes 1997: 44, fig. 4.1).
al. 1991: 218, fig. 9.1; findspots without num- Significantly, the three confirmed prehistoric
bers mark transect scatters outside of identi- settlement sites listed above were all known pre-
fied ‘sites’ of historic or prehistoric date). Yet, vious to the Keos survey, and (other than Ayia
tellingly, they state: Irini) lack evidence of historical occupation,
Virtually none of the sites in the survey area,
while of the remaining 24 prehistoric ceramic
other than Ayia Irini, Kephala, and Paoura, findspots, 18 are either historic sites or findspots
has produced substantial concentrations of where historic sherds outnumber those recog-
prehistoric pottery or other finds. Indeed, at nized of potential prehistoric date. Since off-site
most locations, whether on-site or off-site… transect pottery was not collected systemati-
prehistoric activity is represented by less than cally, and prehistoric sherds in particular were
a handful of pottery and at several by just a extremely rarely observed in fieldwalking (there
single positively diagnostic sherd… Such are only three ‘off-site’ prehistoric pot scatters,
impoverished finds can scarcely be taken as two of which also have historic finds), our view
indicative of the repeated use of a site, let would be that the very high rate of coincidence
alone its seasonal or permanent occupation, of the observed prehistoric pot scatters with the
in prehistoric times (Cherry et al. 1991: 217).
location of historic ‘sites’ is unlikely to be due to
Figure 9. Northern Keos Survey: findspots with prehistoric ceramic or lithic artefacts (Cherry et al. 1991: 218,
fig. 9.1). Unnumbered locations indicate tract, as opposed to on-site, finds.
the repeated selection of a limited number of In the light of our own and others’ experi-
favourable activity locations, or proximity to ence at other urban sites—such as Askra,
coastal settlement nucleations (Cherry et al. mentioned earlier—we would agree with this
1991: 232, n. 1). We propose instead the mech- view, but would also offer a model for the
anism already argued for from the Boeotian Bronze Age thin scatters: we suggest that the
material, whereby the heightened visibility of regular survival of sherds of all Bronze Age
historic pottery sites sets in motion a ‘window’ periods across the later Graeco-Roman urban
for the enhanced detection of far thinner pre- zone points to the site being previously occu-
historic sherd scatters, which are in actuality pied by numerous, small and short-lived resi-
dispersed more generally across the cultivable dential and burial areas, not necessarily ever
landscape. Were such exaggerated attention to representing a large contemporary popula-
surface finds as is given to ‘sites’ to be applied tion. Now since total surface ceramic density
universally to the ‘non-site’ areas of north-west at this site can rise to as much as 5000 sherds
Keos, we believe that a far greater number of per ha, a proportional extrapolation of the
such prehistoric scatters would be revealed. small prehistoric component within the sam-
For the recent Pylos survey, we have already ple dated collection of some 9500 sherds from
noted earlier the potential for a comparable the entire 120 ha site would, as we have
revision of its prehistoric site record. argued for the offsite prehistoric finds in the
Finally, we might mention the urban sur- LSE/THS sector in Boeotia, enrich the plot-
face survey at the site of Phlius (Alcock ted prehistoric scatters at Phlius to a much
1991), conducted as part of the Nemea Valley fuller cover, but of equally thin density, per-
Archaeological Project, primarily to show haps indicating that farm and hamlet occupa-
once more the ubiquity of low-density prehis- tion shifted continuously around the entire
toric sherd and lithic material within rich later surface of the ancient city.
ceramic site carpets of historic age. System-
atic sample collection across the 120 ha
Conclusion
ancient urban site produced some 9500 pieces
of sampled ceramics (Alcock 1991: 443). Just In summary, we have argued in this paper that
10 pieces of prehistoric lithic were recovered; the characteristic triple phenomena (low-fre-
1-2 possible Neolithic sherds (at opposite quency ‘prehistoric site’ distributions with a
ends of the site); 1-2 pieces of EBA pottery strong emphasis on multi-period use and larger
were recorded at 16 locations all over the site size; the common occurrence of small numbers
and at two locations with 2-3 pieces, with just of prehistoric finds on historical sites; and the
one location with more than 3 sherds; MBA chance hints of a landscape draped with
finds amounted to two widely separated loca- extremely low-density carpets of prehistoric
tions with 1-2 sherds; finally LBA finds were potsherds) are indicative of a hitherto unrecog-
recorded at the level of 1-2 sherds from 10 nized prehistoric landscape. A small number of
locations all over the site, at just one with 2- prehistoric settlement sites survive today in
3. Perhaps not surprisingly, Alcock avoids dense surface form, through their unusual size
attempting an interpretation as to what and prolonged period of use (or perhaps occa-
exactly these curious, thin and dispersed dis- sionally, through atypical taphonomic circum-
tributions might represent in terms of prehis- stances having prevented an earlier exposure to
toric settlement, although she does seem to intensive plough damage or natural surface
suggest that the LBA finds (as plentiful as for weathering). But by far the majority of prehis-
any other prehistoric phase) would hardly toric settlements were small sites of farm or
support village status. hamlet character, whose vestigial ceramic
traces are usually not noted at all in intensive About the Authors
survey; if they are, through the double action of
J.L.Bintliff is Professor of Classical Archaeology
rare transect collection and indirect discovery
at the University of Leiden. He was trained as a
within historical site collection, then numeri-
prehistorian at Cambridge by David Clarke and
cally their contribution to surface density is so
E.S. Higgs, and has been successively Lecturer
slight that their one- or two-piece presence on
and Reader in Archaeology at the University of
the surface as recorded attracts no special inter-
Bradford and Reader at the University of
est. In Messenia, as in Boeotia and the other
Durham. His publications include Natural
Greek survey regions discussed in this article,
Environment and Human Settlement in Prehistoric
just as in Bohemia, these sporadic finds of a few
Greece (1977), European Social Evolution:
fragments represent the tip of a giant iceberg of
Archaeological Perspectives (1984), The Annales
many thousands of small and ephemeral occu-
School and Archaeology (1991), Europe Between
pation- and activity-foci, shifting within small
Late Antiquity and the Middle Ages (with H.
areas of landscape, across the millennia of farm-
Hamerow) (1995), and Structure and Contin-
ing prehistory.
gency in the Evolution of Life, Human Evolution
and Human History (1999).
Acknowledgments
The field survey of the Leondari South East and P. Howard is Senior Experimental Officer in
Thespiae South transects was directed by the Archaeology Department at Durham
Anthony Snodgrass and undertaken by the University. He took his BA in Prehistory and
Cambridge University team within the Boeotia Archaeology at Sheffield University and an
Project, a collaboration between Durham (pre- MA in Scientific Methods in Archaeology at
viously Bradford) and Cambridge Universities, Bradford University. His research interests
sponsored by the British Academy and carried include geophysical prospection and archaeo-
out under the joint direction of John Bintliff logical computing, particularly the applica-
and Anthony Snodgrass. The prehistoric tion of Geographical Information Systems.
ceramics were identified by Oliver Dickinson
(Durham University). The computerization of A.M. Snodgrass is Laurence Professor of Classi-
the survey database and its ongoing investiga- cal Archaeology at the University of Cam-
tion utilizing GIS was begun by Mark Gillings bridge. He trained as a Classical Archaeologist
(Leicester University) and is currently the at Oxford University, then was successively
responsibility of Phil Howard (Durham Univer- Lecturer, Reader and Professor at the Univer-
sity). The interpretative analysis of the material sity of Edinburgh, before moving to his present
is the responsibility of John Bintliff and chair in 1976. He has specialized in protohis-
Anthony Snodgrass. This paper has also bene- toric and early historical Greece, and his publi-
fitted from constructive comments by Jeremy cations include The Dark Age of Greece (1971),
Taylor, Chris Mee and Nicola Terrenato, as well Archaic Greece (1980) and An Archaeology of
as those provided by the JMA referees and the Greece (1987).
journal’s editors. Members of the Pylos Regional
Archaeological Project kindly elucidated
References
aspects of their preliminary publications. The
invitation to John Bintliff to undertake a study Alcock, S.E.
visit to the Institute of Archaeology in Prague 1991 Urban survey and the polis of Phlius. Hesperia
was a major catalyst in the making of this paper. 60: 421-63.

Barrett, J.C. 1988b The end of the Roman countryside: a view from
1994 Fragments of Antiquity: An Archaeology of the East. In R.F.J. Jones, J.H.F. Bloemers, S.L.
Social Life in Britain, 2900–1200 BC. Oxford: Dyson and M. Biddle (eds.), First Millennium
Blackwell. Papers:Western Europe in the First Millennium
Bintliff, J.L. AD. BAR International Series 401: 175-217.
1977a Natural Environment and Human Settlement in Oxford: British Archaeological Reports.
Prehistoric Greece. BAR International Series Blackman, D., and K. Branigan
28. Oxford: British Archaeological Reports. 1977 An archaeological survey of the Ayiofarango
1977b New approaches to human geography. Prehis- valley. Annual of the British School at Athens
toric Greece: a case study. In F. Carter (ed.), 72: 13-84.
An Historical Geography of the Balkans, 59- Cavanagh, W.G.
114. London: Academic Press. 1999 Revenons à nos moutons: surface survey and
1985 The Boeotia Survey, Central Greece. In S. the Peloponnese in the Late and Final
Macready and F.H. Thompson (eds.), Archae- Neolithic. In J. Renard (ed.), Le Péloponnèse:
ological Field Survey in Britain and Abroad, 196- Archéologie et histoire, 31-65. Rennes: Presses:
216. London: Society of Antiquaries. Universitaires Rennes.
1991 The Roman countryside in Central Greece: Chapman, J., R. Shiel and S. Batovic
observations and theories from the Boeotia 1996 The Changing Face of Dalmatia. Archaeological
Survey (1978–1987). In G. Barker and J. and Ecological Studies in a Mediterranean Land-
Lloyd (eds.), Roman Landscapes: Archaeologi- scape. London: Cassell and Leicester Univer-
cal Survey in the Mediterranean Region, 122-32. sity Press.
London: British School at Rome. Carothers, J., and W.A. McDonald
1996 The archaeological survey of the Valley of the 1979 Size and distribution of the population in
Muses and its significance for Boeotian History. Late Bronze Age Messenia: some statistical
In A. Hurst and A. Schachter (eds.), La Mon- approaches. Journal of Field Archaeology 7:
tagne des Muses, 193-224. Geneva: Librairie Droz. 433-54.
1997a The archaeological investigation of deserted Cherry, J.F.
medieval and post-medieval villages in 1979 Four problems in Cycladic prehistory. In J.L.
Greece. In G. De Boe and F. Verhaeghe Davis and J.F. Cherry (eds.), Papers in Cycladic
(eds.), Rural Settlements in Medieval Europe. Prehistory. Monograph 14: 22-47. Los Ange-
Papers of the ‘Medieval Europe Brugge 1997’ les: UCLA Institute of Archaeology.
Conference, vol. 6: 21-34. Bruges: Zellik. 1982 A preliminary definition of site distribution
1997b Further considerations on the population of on Melos. In C. Renfrew and M. Wagstaff
ancient Boeotia. In J.L. Bintliff (ed.), Recent (eds.), An Island Polity: The Archaeology of
Developments in the History and Archaeology of Exploitation in Melos, 10-23. Cambridge:
Central Greece. BAR International Series Cambridge University Press.
666: 231-52. Oxford: Tempus Reparatum. 1983 Frogs round the pond: perspectives on current
1997c Regional survey, demography, and the rise of archaeological survey projects in the Mediter-
complex societies in the ancient Aegean: core- ranean region. In D.R. Keller and D.W. Rupp
periphery, neo-malthusian, and other interpre- (eds.), Archaeological Survey in the Mediterranean
tive models. Journal of Field Archaeology 24: Area. BAR International Series 155: 375-416.
1-38. Oxford: British Archaeological Reports.
Bintliff, J.L., and A.M. Snodgrass 1984 Common sense in Mediterranean survey?
1985 The Boeotia survey, a preliminary report: the Journal of Field Archaeology 11: 117-20.
first four years. Journal of Field Archaeology 12: Cherry, J.F., J.L. Davis and E. Mantzourani
123-61. 1991 Landscape Archaeology as Long-Term History:
1988a Off-site pottery distributions: a regional and Northern Keos in the Cycladic Islands. Monu-
interregional perspective. Current Anthropol- menta Archaeologica 16. Los Angeles:
ogy 29: 506-13. UCLA Institute of Archaeology.

Clark, R.A., and A.J. Schofield field survey in Bohemia. Pamatky Archeolog-
1991 By experiment and calibration: an integrated icke 84(2): 110-30.
approach to archaeology of the ploughsoil. In McDonald, T.
A.J. Schofield (ed.), Interpreting Artefact Scat- 1993 The A41 Excavations. Current Archaeology
ters: Contributions to Ploughsoil Archaeology, 136: 133-37.
93-105. Oxford: Oxbow. McDonald, W.A.
Davis, J.L., S.E. Alcock, J. Bennet, Y.G. Lolos and C.W. 1984 The Minnesota Messenia Survey: a look
Shelmerdine back. In A.L. Boegehold et al. (eds.), Studies
1997 The Pylos Regional Archaeological Project. Presented to Sterling Dow. Greek, Roman, and
Part I: Overview and the archaeological sur- Byzantine Studies Monograph 10: 185-91.
vey. Hesperia 66: 391-494. Durham, NC: Duke University Press.
Davis, J.L. (ed.) McDonald, W.A., and G.R. Rapp, Jr (eds.)
1998 Sandy Pylos: An Archaeological History from 1972 The Minnesota Messenia Expedition: Recon-
Nestor to Navarino. Austin: University of Texas structing a Bronze Age Regional Environment.
Press. Minneapolis: University of Minnesota Press.
Di Gennaro, F., and S. Stoddart Mee, C., and H. Forbes (eds.)
1982 A review of the evidence for prehistoric 1997 A Rough and Rocky Place: The Landscape and Set-
activity in part of South Etruria. Papers of the tlement History of the Methana Peninsula, Greece.
British School at Rome 50: 1-21. Liverpool Monographs in Archaeology and Ori-
Hope Simpson, R. ental Studies. Liverpool: Liverpool University
1984 The analysis of data from surface surveys. Press.
Journal of Field Archaeology 11: 115-17. Millett, M.
1985 The evaluation of data from surface surveys. 1991 Pottery: population or supply patterns? In
Journal of Field Archaeology 12: 258-60. G. Barker and J. Lloyd (eds.), Roman Land-
Jacobsen, T.W. scapes: Archaeological Survey in the Mediterranean
1984 Seasonal pastoralism in Southern Greece: a Region, 18-26. London: British School at Rome.
consideration of the ecology of Neolithic Neustupny, E.
Urfirnis pottery. In P.M. Rice (ed.), Pots and 1998 The transformation of community areas into
Potters: Current Approaches in Ceramic Archae- settlement areas. In E. Neustupny (ed.), Space
ology, 27-43. Los Angeles: UCLA Institute of in Prehistoric Bohemia, 45-61. Prague: Institute
Archaeology. of Archaeology.
Jameson, M.H., C.N. Runnels and T.H. van Andel Pope, K.O, and T.H. van Andel
1994 A Greek Countryside: The Southern Argolid 1984 Late Quaternary alluviation and soil forma-
from Prehistory to the Present Day. Stanford: tion in the southern Argolid. Journal of
Stanford University Press. Archaeological Science 11: 281-306.
Kuna, M. Renfrew, C., and M. Wagstaff (eds.).
1991 The structuring of prehistoric landscape. 1982 An Island Polity: The Archaeology of Exploita-
Antiquity 65: 332-47. tion in Melos. Cambridge: Cambridge Univer-
1998 The memory of landscapes. In E. Neustupny sity Press.
(ed.), Space in Prehistoric Bohemia, 106-115. Reynolds, P.
Prague: Institute of Archaeology. 1982 The ploughzone. In Anon (ed.), Festschrift zum
In press Surface artefact studies in the Czech Republic. 100 jährigen Jubiläum der Abteilung Vorgeschichte
In J. Bintliff, M. Kuna and N.Venclova (eds.), der naturhistorischen Gesellschaft Nürnberg, 315-
The Future of Archaeological Field Survey in 40. Nuremberg.
Europe. Sheffield: Sheffield Academic Press. Richards, J. (ed.).
Kuna, M., M. Zvelebil, P.J. Foster, and D. Dreslerova 1990 The Stonehenge Environs Project. London: His-
1993 Field survey and landscape archaeology toric Buildings and Monuments Commission
research design: methodology of a regional for England.

Runnels, C.N. Bronze Age of the southern and central
1982 Flaked-stone artifacts in Greece during the Greek Mainland. American Journal of Archae-
historical period. Journal of Field Archaeology ology 101: 537-85.
9: 363-73. Snodgrass, A.M.
Rutter, J.B. 1987 An Archaeology of Greece. Berkeley and Los
1983 Some thoughts on the analysis of ceramic Angeles: University of California Press.
data generated by site surveys. In D.R. Keller Wagstaff, M., and J.F. Cherry
and D.W. Rupp (eds.), Archaeological Survey 1982 Settlement and population. In C. Renfrew
in the Mediterranean Area. BAR International and M. Wagstaff (eds.), An Island Polity: The
Series 155: 137-42. Oxford: British Archaeo- Archaeology of Exploitation on Melos, 136-55.
logical Reports. Cambridge: Cambridge University Press.
Salac, V. Whitelaw, T.M.
1995 The density of archaeological finds in settle- 1991 Investigations at the Neolithic sites of
ment features of the La Tène period. In Kephala and Paoura. In J.F. Cherry, J.L. Davis
M. Kuna and N. Venclova (eds.), Whither and E. Mantzourani, Landscape Archaeology as
Archaeology? Papers in Honour of Evzen Long-Term History: Northern Keos in the
Neustupny, 264-76. Prague: Institute of Cycladic Islands. Monumenta Archaeologica
Archaeology. 16: 199-216. Los Angeles: UCLA Institute of
Schofield, A.J. Archaeology.
1987 The role of palaeoecology in understanding Zangger, E., M.E. Timpson, S.B. Yazvenko, F. Kuhnke
variations in regional survey data. Circaea 5: and J.Knauss
33-42. 1997 The Pylos Regional Survey Project. Part II:
Shelmerdine, C.W. landscape evolution and site preservation.
1997 Review of Aegean Prehistory VI: The Palatial Hesperia 66: 549-641.

The Hidden Landscape of Prehistoric Greece - Bintliff - Howard &amp Snodgrass

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Hidden Landscape of Prehistoric Greece - Bintliff - Howard &amp Snodgrass

Uploaded by

Copyright:

Available Formats

01 Bintliff et al.

16/11/01 3:19 pm Page 139

Journal of Mediterranean Archaeology 12.2 (1999) 139-168

John Bintliff1, Phil Howard2 and Anthony Snodgrass3

Introduction sites and ‘offsite’ finds, in which the artefacts of

Bintliff, Howard and Snodgrass

The Hidden Landscape of Prehistoric Greece 141

142 Bintliff, Howard and Snodgrass

throughout the Bronze Age. Moreover, a walking characteristic of these surveys

The Hidden Landscape of Prehistoric Greece 143

144 Bintliff, Howard and Snodgrass

The Hidden Landscape of Prehistoric Greece 145

146 Bintliff, Howard and Snodgrass

The Hidden Landscape of Prehistoric Greece 147

148 Bintliff, Howard and Snodgrass

The Hidden Landscape of Prehistoric Greece 149

150 Bintliff, Howard and Snodgrass

The Hidden Landscape of Prehistoric Greece 151

152 Bintliff, Howard and Snodgrass

The Hidden Landscape of Prehistoric Greece 153

154 Bintliff, Howard and Snodgrass

The Hidden Landscape of Prehistoric Greece 155

156 Bintliff, Howard and Snodgrass

The Hidden Landscape of Prehistoric Greece 157

158 Bintliff, Howard and Snodgrass

The Hidden Landscape of Prehistoric Greece 159

160 Bintliff, Howard and Snodgrass

The Hidden Landscape of Prehistoric Greece 161

162 Bintliff, Howard and Snodgrass

The Hidden Landscape of Prehistoric Greece 163

164 Bintliff, Howard and Snodgrass

The Hidden Landscape of Prehistoric Greece 165

166 Bintliff, Howard and Snodgrass

The Hidden Landscape of Prehistoric Greece 167

168 Bintliff, Howard and Snodgrass

You might also like