Thursday, August 19, 2010

The final (?) word on those handaxes from Crete

While everybody was busy talking about unexpectedly old cutmarks and other Pleistocene goings-on last week, the paper by Strasser et al. (2010) describing the discovery of quartz handaxe assemblages on Crete ResearchBlogging.orgquietly came out in Hesperia. This is a topic that was discussed at length on this blog, in several posts that generated a large amount of comments a few months back. The sticking point of all the arguments concerned the chronology of the handaxes, so without further ado, here's the money quote (Strasser et al. 2010: 185-186):

"The dating of the Palaeolithic in the Plakias region presents a considerable challenge, not least because of the long period of time that may have elapsed since the occupation of the earliest sites, during which postdepositional natural processes may have obscured the archaeological record. Additionally complicating the issue are the small number of sites, the lack of excavation, and the impact of modern development on the area, which has destroyed many sites.

Several approaches to dating were attempted, and our research on this topic continues. At Preveli 2, east of the Preveli Gorge, Palaeolithic artifacts are associated with a flight of marine terraces resulting from relatively high sea levels in the Pleistocene that were preserved by subsequent rock uplift. The lowest late Pleistocene marine terraces resulting from high stands of the sea at Preveli (14 ± 1 masl) and Schinaria (21 ± 1 masl) have 2-sigma calibrated radiocarbon ages of 45,400 ± 1,600 and 49,120 ± 2,890 years b.p., respectively, and are correlated with Marine Isotope Stages 3.3 and 3.4, both eustatic high stands. The higher terraces, at 59 and 96 masl, are unquestionably older. How much older? Assuming similar rates of rock uplift (1.4 ± 0.1 m/kyr) determined from the age-elevation relationships of the dated terraces at 14 and 21 masl, it is possible to estimate the approximate ages of the terraces associated with artifacts. This correlation provides an approximate age for the lithic artifacts. The higher terrace, at 96 masl, may belong to Marine Isotope Stage 5, possibly early 5e, ca. 110,000 b.p. Artifacts associated with the terrace at 59 masl could correlate with Marine Isotope Stage 5a, ca. 70,000 b.p. It should be stressed that these are rough approximations and these ages are probably minima that represent a terminus ante quem. If the uplift rate is changed, the terraces and the artifacts associated with them could be much older.

At Preveli 3, Preveli 7, Timeos Stavros 1, and Schinaria 5, Palaeolithic artifacts were found in outcrops of paleosols that exhibit the characteristics of the oldest maturity stage for such features, that is, Maturity Stage 6, or in geological terms, Marine Isotope Stage 6. Together these observations suggest an age of ca. 190,000–130,000 b.p. and serve as a terminus ante quem for the artifacts embedded within them. The stone tools were incorporated in the paleosols as part of a process described by Runnels and van Andel in Epirus: “the top of the Bt horizon itself would move gradually upward as a result of slow deposition, so engulfing any artifacts laid down on former land surfaces above it.” In other words, the Bt horizon, especially as much of the clay comes from eolian sources, will increase in thickness through time, slowly
engulfing clasts, such as stone tools, that were formerly in the A horizon.

In sum, the dating of the Palaeolithic sites is based on geological data derived from the study of marine terraces on the southwestern coast of Crete and our identification of paleosols, and these data place the Palaeolithic lithic artifacts firmly in the Pleistocene, ca. 130,000 b.p. or earlier. The chronology can be further refined, however, and a dating program
currently in progress may provide data for doing so." (references excised)

So, bottom line, the dating is largely indirect, but grounded by dated references points that provide a minimum age for the terraces where the handaxes were recovered. Interestingly, this indicates that these tools are, at most, 130,000 years old. Given that there do not appear to be more recent Paleolithic age (the rest of the implements reported in the paper are Mesolithic in age and techno-typology, which is an important discovery in and of itself), this implies that on current evidence, Crete was not occupied during the Late Pleistocene (ca. 130,000-10,000 BP). Why this was the case (and how an early colonization took place) is an interesting question that will need to be answered by future research.

In the meantime, here are a few take-away observations from the Strasser et al. (2010) report. First, on the basis of the drawing of the handaxes, these implements do appear to be human-made. Second, they are not isolated occurrences: the authors identified nine localities where these quartz tools were found, only three of which also yielded Mesolithic tools. This leaves open the possibility that the 'Paleolithic' sites represent task-specific components of the Mesolithic toolkit on Crete, but this is unlikely based on the association of handaxes with some of the terrace deposits described in the quote above. Third, as the authors indicate, this was not a case of a H. heidelbergensis (or a couple of them) washing onto Crete: the fact that nine sites (defined by the presence of a minimum of 20 stone tools) were found in a relatively small area indicates a somewhat sustained human presence on the southern coast of Crete. This does suggest that people got there purposefully (i.e., using some kind of watercraft), which leaves open the question of why Middle and Upper Paleolithic assemblages haven't (yet?) been found in the region or elsewhere on Crete (or any other large Mediterranean island, for that matter).

As a parting observation: one aspect of the discovery that definitely wasn't stressed in the media reports about these finds is how they were made. They were breathlessly reported as 'discovered' without much context. The authors actually set out to look for Mesolithic sites in southern Crete using a model developed for the Greek mainland that identified certain areas as having been most appealing for Mesolithic foragers. In other words, this wasn't a blind search for early stuff or simply a fortuitous discovery. Rather, it came about as the result of an explicit research design targeting some very specific questions. Good to see more of that in Paleolithic archaeology. 

References:

Strasser, T., Panagopoulou, E., Runnels, C., Murray, P., Thompson, N., Karkanas, P., McCoy, F., & Wegmann, K. (2010). Stone Age Seafaring in the Mediterranean: Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Crete Hesperia, 79 (2), 145-190 DOI: 10.2972/hesp.79.2.145



26 comments:

terryt said...

"this implies that on current evidence, Crete was not occupied during the Late Pleistocene (ca. 130,000-10,000 BP)".

So much for the prolonged period of occupation Maju claimed.

"Why this was the case (and how an early colonization took place) is an interesting question that will need to be answered by future research".

True. But presumably the population became isolated and inbred, and died out.

"This does suggest that people got there purposefully (i.e., using some kind of watercraft), which leaves open the question of why Middle and Upper Paleolithic assemblages haven't (yet?) been found in the region or elsewhere on Crete (or any other large Mediterranean island, for that matter)".

Yes. Why have they not been found elsewhere? Suggests that watercraft were not the method by which these people reached Crete otherwise, surely, they would have reached other islands too. Tyhe region is tectonically active so perhaps some land bridge formed for a short time.

Maju said...

Let's see, Julien. You say that:

"... this indicates that these tools are, at most, 130,000 years old".

But in the quoted text we see dates ranging from 45 Ka to 130-190 Ka., including also mentions to 70 Ka and 110 Ka sites (fossil beaches with artifacts embedded in them). So the earlier conclusions of 130-40 Ka for this quartz industry stand more or less (some precisions added but nothing substantial changed), with a more precise range of 190-45 Ka. as terminus ante quem (i.e. at or before those dates, as they represent only the deposit of the tools in the beaches).

Julien Riel-Salvatore said...

Terry - I'd be very surprised if tectonic activity created a short-lived land bridge that hasn't been picked up on bathymetric maps... if anything, I'd expect that the know-how involved in boat-building might have been lost in a small population over time. There are some interesting archaeo/ethnographically documented analogs of such processes taking place in other island populations.

Julien Riel-Salvatore said...

Maju -
that's incorrect: 45 kya is the age of uplifted terraces that do not contain artifacts. It's by looking at how much these artifact-free terraces have been uplifted and generalizing that uplift rate to other, older terraces that did yield artifacts that Strasser et al. are able to propose tentative dates, the very earliest of which might be 70 kya (for one of the nine localities they identified). The authors emphatically state that these are only minimum ages, and that the actual age is likely to be much older. For the other five sites for which they can evaluate the age based on similar calculations, the minimum age of the artifact-bearing terraces in between 190-130 kya. So, while there is a small possibility that one locality may dates to the end of OIS 5, the bulk of the evidence suggest that they all date to OIS 6 at the earliest. Thus, parsimony suggests that the evidence presented by Strasser and co. is bets interpreted as showing that Crete was devoid of human occupation after about 130kya.

Maju said...

That would change everything in relation to the previous release.

Still I do see a clear reference to the 70 Ka years beach and the 110 Ka one, so I cannot conclude as you do that 130Ka is the most recent possible date. The logical conclusion would be if anything that 70 Ka is the only possible date.

When the authors say: "In sum (...) these data place the Palaeolithic lithic artifacts firmly in the Pleistocene, ca. 130,000 b.p. or earlier".

They they must mean (from context) that the earliest human habitation of Crete happened 130 Ka or earlier. But that says nothing about what happened next.

CamArchGrad said...

Argg.. a log in access.

I'd like to read the paper before I comment, as it seems they have really taken the time to do a detailed study.

Do you have a copy?

Julien Riel-Salvatore said...

CAG -
yes, I have a copy; write to me and I'll be glad to send you one as well (this goes for anyone of the thread).

JRS

Julien Riel-Salvatore said...

Maju -
well, again, the authors are treating the ages they propose as absolute minima, meaning the deposits are in all likelihood older. And, again, the earliest ages they propose are for a single locality - if we work on the assumption that the handaxe-bearing deposits in the region are more or less coeval, then it makes sense that the minimum of 130-190kya which they estimate for most of the other localities can be extended to the one where their rough estimates proposes a 70-110kya terminus ante quem.

In any case, even if the 70kya determination turned out to be correct the question remains the same: what happened to people on Crete b/w 70-10kya? That's the real question here.

terryt said...

"what happened to people on Crete b/w 70-10kya? That's the real question here".

Died out through inbreeding, caused by lack of new immigrants, would be my first guess.

Maju said...

Reading now the paper, it is somewhat confusing regarding the chronology. It is clear that the authors are quite safe and emphatic about the 130-190 Ka BP date for a minimum "terminus ante quem" age of arrival (i.e. at least before 130 Ka) but the duration of the population/industry is less clearly described.

Early on they mention a 19th century finding of skeletal remains at Chania (species?) with the rock layer dated to 63-39 Ka BP. They also mention findings by another team in Gavdos island, south of Crete, "belonging to all phases of the Palaeolithic".

The "Chronology" section that you quoted almost in full is not too clarifying. While the 190-130 Ka BP is clear, sentences like this one:

"Artifacts associated with the terrace at 59 masl could correlate with Marine Isotope Stage 5a, ca. 70,000 b.p."

Are clearly indicative of Paleolithic artifacts potentially more recent than 130 Ka. 49 and 45 Ka are also mentioned but, confusingly, no explicit mention of artifacts is made in this case and I cannot extrapolate in either sense from the rest of the paper.

It seems that the research (and publication - at least two unpublished materials are referred to in the text) on Paleolithic Crete is still very much open.

What is clear is that we have paleo-navigators able to reach Crete, presumably by "island-hoping" from either Greece or Turkey, before 130 Ka., what is in itself the oldest direct evidence of such daring endeavor AFAIK, older than H. floresiensis (94 Ka) and certainly older than the metatarsal from Philippines (63 Ka).

CamArchGrad said...

As much as I'd like to say yes to the dress, I'm going to have to say no.

First thanks to Julian for sending me the paper. I had a look but a couple things looked wrong and I couldn't quite place them. So I looked at http://misliya.haifa.ac.il/archaeology/lithic/lithic.html"
and http://web.arizona.edu/~hatayup/pubs/Kuhn2002.pdf and http://antiquity.ac.uk/projgall/vermeersch/index.html Egypt

And it came together. They don't look at all like surrounding Middle Paleolithic assemblages. As can be seen from the links above, the industries surrounding Crete, were composed of retouched levallois and blade industries.

these crude artifacts looked nothing like what someone one would have created, even expediently. So I looked at the geo-archeological setting, and went uh-oh. There are large blocks of quartz rolling down hill (fig 14) and spalling away and have been doing so since time immemorial?? Looking at the contexts, there are literally thousands of shattered rocks (not quartz) but of banged up nevertheless. Also I'm disputing their place in the terrace. If they are the surface they could have come down in a 1950's rainstorm and no-one would be the wiser (fig 31 suggests this).

In a tectonically active environment it's not enough to go to a debris field and select the most likely. The tools in Photo (the drawings are much more tool like) are crude, barely retouched cobbles with little or no "forming". Even as expedient tools I'd be hard pressed. It reminds me all to much of this http://pubs.nrc-cnrc.gc.ca/rp/rppdf/e00-124.pdf

pt 1

CamArchGrad said...

pt 2

It's all too easy to see how those tools got created. Quartz boulders roll down hill leaving a trail of flakes, chunks and other debris. Earthquakes and erosion and megafuana kick and beat the fragments. At one stage they get buried (and perhaps flaked by the weight of the overlying soil al la Eoliths) maybe buried multiple times and then weathered out again. In terrains of siliceous rocks this is very common, I can go to the creek outside my house and find oldowan material if I wish
(I live in Canada btw). So now we have this bashed quartz block thats been tumbled around for 1 million years (remember the recovery site is only their latest stop, these process could have been operating for millions of years beforehand). It falls in that range of variation between human and natural just waiting for an archaeologist to come along and select it. " it is difficult at times to recognize the conchoidal fractures and the remains of flake removals; in some cases they were hardly visible, making it necessary to trace them with our fingers under various conditions of raking light in order to draw them" Indeed. Or perhaps self-creating them.

Is it too much to ask for a nice site with fauna, features and lithics in situ? In a solid silt/sand matrix? where the tools look like their counterparts elsewhere?

This stands in stark contrast to their Mesolithic material which shows a range of variation in lithic source (chert and quartz), morphology, is definitely formed and retouched (see fig 15 & 16) and looks like surrounding assemblages (Franchiti, Cyprus, Epirus).

Until better evidence comes along, I don't think we have paleolithic seafarers yet.

There are several tests I would like to see done. - Excavation of a cave site. The occupation is synchronous with Skulh and Quafzeh and Haua Ftear, if they visited crete for any length of time there should be cave site or two in the region.

- Nice stone tools. Not bashed up dubious ones, drawn so they look like tools. From turkey to Libya there was blade and levallois technology. Crete has lots of chert so there is no reason for that technology not to exist (how about some lithic reduction sites, with nice refitting flakes as well?).

- Faunal turnover. Humans are indiscriminate predators and devastate previously hermetic environments. Hawks mentions a big turnover at 300k and the next is at when the Mesolithic arrives. Is there any evidence for one around the time of these tools?

- And finally but not really hoping, a hominid. What broke the back of resistance to pre 500k inhabitation of Europe was Atapuerca (Gran Dolina) and Ceprano two undeniable hominid fossils that came from solid context.

terryt said...

"What is clear is that we have paleo-navigators able to reach Crete, presumably by 'island-hoping' from either Greece or Turkey, before 130 Ka., what is in itself the oldest direct evidence of such daring endeavor AFAIK"

Sorry. Probably not. And as CamArchGrad said, 'Is there any evidence for [faunal turnover] around the time of these tools?' More evidenc eagainst ancient human presence on Crete. And the difficulty remains that if they really were capable of 'island-hoping' how come the only island they hopped to was Crete? No others. And there could be another explanation for the artifacts:

"There are large blocks of quartz rolling down hill (fig 14) and spalling away and have been doing so since time immemorial?? Looking at the contexts, there are literally thousands of shattered rocks (not quartz) but of banged up nevertheless".

So the hand-axes could be natural features?

"Also I'm disputing their place in the terrace. If they are the surface they could have come down in a 1950's rainstorm and no-one would be the wiser (fig 31 suggests this)".

I too had difficulty in accepting the authors' dates. They seemed to be basing their dating on the time of terraces' rising, yet obviously the artifacts could have been deposited almost as recently as the day before they were found. The only real justification for the dating appears to be that 'Palaeolithic artifacts were found in outcrops of paleosols that exhibit the characteristics of the oldest maturity stage for such features, that is, Maturity Stage 6, or in geological terms, Marine Isotope Stage 6'.

"Until better evidence comes along, I don't think we have paleolithic seafarers yet".

I totally agree. Certainly not in the Mediterranean anyway.

Maju said...

Let's see: I am not the archaeological expert here but I cannot accept that all those well defined "acheuleoid" artifacts are the product of random erosion. Quartz is very hard and erosion is most unlikely to produce such shapes in any rock I can think of.

"So the hand-axes could be natural features?"

My opinion is: no way. It may happen randomly in one or few rocks, I guess, but the collection is huge. They do not look like the usual pebbles and boulders you normally get that way and instead they do look very clearly like stone tools.

"And the difficulty remains that if they really were capable of 'island-hoping' how come the only island they hopped to was Crete? No others".

The question may be redefined as "how come archaeologists have not found so far similar artifacts in other islands, for instance Cyprus?" They may have arrived elsewhere but the evidence is yet to be found, right?

CamArchGrad said...

Maju,

I would encourage you to look at http://pubs.nrc-cnrc.gc.ca/rp/rppdf/e00-124.pdf.

The fact that siliceous rocks, cherts, quartz, silex, basalts, fine grained rhyolites et al can and will produce through various natural means crude tools has been known since 1914. SH Warren produced the pioneering study on the mechanics (reference available in linked pdf), the onus has been on the describer to prove the tools are human, by other means rather than simple typology.

Don't be fooled by the drawings- they are highly interpretive. The photos tell a much different story.

To repeat, they look nothing like contemporaneous assemblages (From Haua Ftear to Skuhl, to Karian in turkey) see( ffzg.hr/arheo/ska/tekstovi/paleolithic_levant.pdf )

They are randomly associated without context (no refits, no fauna, no features, no hominid remains)

There are legitimate reasons to believe natural processes created them

There is no reason to treat them as anything more vaguely tool like geofacts and no representative of anything more than natural weathering processes.

Maju said...

Driver argues against "cobbles" similar to Oldowan (aka pebble culture, aka chopper industry), something much less elaborate than the Acheulean we are discussing here. Also Driver argues for glacial production of the fractures on such cobbles, something that is not at play in Crete.

You may have been mislead by the lack of images in Driver's paper.

These Acheulean-style artifacts are not at all like pebble/cobble industries or naturally fractured pebbles/cobbles. They are not even rounded by natural erosion and show instead the points and edges proper only of human manufacture, specially in such amounts.

The difference of industry style with nearby areas of West Eurasia was also noticed in the earlier press release and actually they suggested African affinities (prior to MSA, of course). This is no argument anyhow because we cannot even hope to know of most actual sites even in the well-surveyed areas, much less for that time depth. If the arrival happened, say, 200,000 years ago, neither Mousterian nor other industries would have arrived to the area yet probably.

Even if there is some room for caution, as always happens with anomalous findings, your arguments do not seem to amount to a clear falsation of the findings (not even close).

terryt said...

Connected to a discussion concerning mammoth extinction I've been having with Maju at his blog. I found this most interesting particular case, the Mediterranean islands:

http://en.wikipedia.org/wiki/Dwarf_elephant

Technically the only 'mammoth' is the species on Sardinia, but:

"Fossil remains of dwarf elephants have been found on the Mediterranean islands of Cyprus, Malta (at Ghar Dalam), Crete, Sicily, Sardinia, the Cyclades Islands and the Dodecanese Islands".

And:

"The Cyprus dwarf elephant survived at least until 11,000 BP".

And, particularly interesting, concerning the Dodecanese Islands:

"They became extinct just less than 4,000 years BP, so this elephant survived well into the Holocene".

This provides supporting evidence that no significant boating ability appeared in the Mediterranean until around 10,000 years ago. So, if the artifacts on Crete are human-made how did the humans get there?

Maju said...

Survival of the elephants proves nothing, Terry, specially as Cyprus had been inhabited for many millennia before these animals went extinct.

Julien Riel-Salvatore said...

OK - back on this after a few days.

CAG, I can only agree with you that what is really needed to put the issue to rest is an in situ assemblage, ideally with refitting pieces that can show the exterior platform angles and any platform preparation employed in detaching flakes. Until then, the age remains conjectural. And It certainly is odd that the Cretan tools are so typologically and technologically distinct from anything else in southern Europe or the Levant at that time.

As for whether or not they're geofacts/eoliths, while I agree that drawings are interpretative, I'd be surprised that people who have looked at lithics in the past (as co-author C. Runnels certainly has) would not have considered this. Then again, R. Leakey was fooled by the Calico finds that turned out to be natural, so who knows, but still... but again, this is a question that would be resolved by finding artifacts in stratigraphic context, which is presumably a next step in their research program. In any case, it should be easy for the authors to conduct a study comparing naturally broken pieces of quartz to their assemblage to see whether or not they have some real on their hand, as Gillepsie et al. (2004) have done for alleged pre-Clovis assemblages in Alberta.

CamArchGrad said...

OH! That assemblage is still around. People were dismissing that in 1997. But it's an interesting paper, nonetheless.

It's a classic example of science hypothesis and vigorous testing to confirm or deny.

terryt said...

"specially as Cyprus had been inhabited for many millennia before these animals went extinct".

Really?

terryt said...

Sorry. Me again.

"specially as Cyprus had been inhabited for many millennia before these animals went extinct".


Then I'm sure you'll find this very interesting:

http://www.bu.edu/asor/pubs/books-monographs/swiny.pdf

Maju said...

Terry: nice book but from 10 Ka (first known human in Cyprus) to 4 Ka (elephant extinction) it is many millennia, right? The dodo did not last even a fraction of that.

terryt said...

"from 10 Ka (first known human in Cyprus) to 4 Ka (elephant extinction) it is many millennia, right?"

True. But irelevant. I've mentioned before that you don't actually read any links I provide. The extinction happened at different times on different islands. So I'll repeat myself:

"The Cyprus dwarf elephant survived at least until 11,000 BP".

So the time gap between '11,000 BP' and '10 Ka' from the'first known human in Cyprus' and elephant extinction on that island is reasonably close to the lenghth of time it took for the demise of the dodo. Your '4 Ka' date for elephant extinction has no relationship whatsoever to Cyprus. But:

"concerning the Dodecanese Islands: 'They became extinct just less than 4,000 years BP'"

The time it took for the elephants to become extinct on the Dodecanese Islands is unknown, but it is my guess their extinction on those islands was also relatively rapid. The surprising thing about that is it suggests those island were settled long after most other Mediterranean islands.

Maju said...

Sure, sorry. I somehow got the idea that the elephants were not extinct before 4000 BP, when it's been all the time 11 Ka. My bad.

terryt said...

Thank you. My last link was actually in relation to your claim 'Cyprus had been inhabited for many millennia before these animals went extinct'. The link clearly debunks that idea (page 14):

"Despite such claims, Cherry has convincingly demonstrated that there are few, if any, compelling arguments for people being on most of the Mediterranean islands during the Late Pleistocene or Easrly Holocene".

I am reasonably certain that a list of dates for elephant (and other) extinction on the various Mediterranean islands will give us a very good indication of the time and pattern of human dispersal through those islands. I look forward to reading someones research on the matter.