Reader Comments
Post a new comment on this article
Post Your Discussion Comment
Please follow our guidelines for comments and review our competing interests policy. Comments that do not conform to our guidelines will be promptly removed and the user account disabled. The following must be avoided:
- Remarks that could be interpreted as allegations of misconduct
- Unsupported assertions or statements
- Inflammatory or insulting language
Thank You!
Thank you for taking the time to flag this posting; we review flagged postings on a regular basis.
closeCaution is Needed, But Not in the Manner Indicated by the Authors
Posted by DwightRead on 02 Jan 2014 at 06:14 GMT
The experimental results presented by the authors say less about the assumption that “similarity of shape is … an indicator of cultural … similarity” than the authors believe. The rationale for this assumption lies in artisans sharing the same target artifact shape and then making similarly shaped artifacts. Similarity becomes an index of cultural context to the extent that the artisans share the same target shape for cultural reasons. However, while "culturally shared target shape" → "similarity of outcome", the converse is not true in general. Were we to find similarly shaped artifacts both in an archaeological site in France and in India, we would not conclude from that fact alone cultural identity between the two contexts. We would first need to establish whether such a claim is even plausible: Do the two contexts come from the same time period? Was there trade between the two regions? And so on.
What the experiment does show is the validity of the assumption that when artisans have the same target in mind, they produce similar artifacts. In this case, it is the experimenters, rather than shared cultural context, that introduces the similarity of target since the experimenters asked both sets of potters to reproduce the same spherical-shaped pot. That the authors find similarity in the outcomes of the potters, regardless of their locality, simply underscores the validity of the (modified) implication: "shared target shape" → "similarity of outcome", where we drop the modifier "culturally" since the source of the similarity of the target shape is now the experiment. The source of the target is not the critical factor, only that the potters share the same shape as the target for their pottery production.
Rather than introducing caution as the authors suggest, the experiment actually provides further evidence for the validity of the assumption "culturally shared target shape" → "similarity of outcome". What they have shown experimentally is that even potters from different contexts make similar artifacts when they share the same target for their artifact production. Where caution "is" needed lies in not making the converse fallacy, namely that from the validity of the assumption we incorrectly assume that the converse, "similarity of outcome" → "culturally shared target shape", is also valid. That the converse is not valid, in general, is trivially true and hardly needs verification.
If our goal is to discern culturally salient types, we do not begin with a data set composed of artifacts from a site in France and a site in India; rather, because initially we do not know the temporal and spatial boundaries of the, as yet, not discerned types we are trying to discover, we begin by “initially restricting data sets to contexts, such as a single assemblage or a distinctive portion of an assemblage, under the assumption that a single assemblage is contained within, and does not straddle, a culturally salient space and time boundary” [1, p. 305]. Thus we will first separate our data into two subsets: artifacts from a site in France and artifacts from a site in India, and conduct our analysis on each subset separately. As a result, our quantitative analysis might lead us (assuming now that we are dealing with spherical shaped pots that have actually been found in an archaeological site in France and in a site in India) to conclude that spherical shaped pots are culturally salient in the French context, and, independently, spherical shaped pots are culturally salient in the Indian context. We might further investigate whether the two instances of spherical shaped pots are independent of each other, or has there been trade, and so on, but we would not conclude that spherical shaped pots are a single cultural type that somehow is present in these two, widely separated regions merely because they have the same shape, which is what the authors assume would be the conclusion derived from a typological classification. This would not, however, be the conclusion derived from a properly done typological classification.
"'References'"
[1] Read DW (2009) Artifact Classification: a Conceptual and Methodological Approach. Left Coast Press.
RE: Caution is Needed, But Not in the Manner Indicated by the Authors
Enora replied to DwightRead on 01 Apr 2014 at 06:59 GMT
We read carefully the thoughtful comments of Prof. D. W. Read and acknowledge him for the scrutiny reading of our article.
We agree that our assumption that “similarity of shape is an indicator of cultural similarity” can be interpreted as simplistic. This assumption could have been moderated in our text underlying – as you suggest – the fact that archaeologists usually take into account various aspects of ceramics to identify their cultural origins. In particular, as Prof. Read noticed region and time period are the first aspects that guide the analysis. However, we have the impression that too much emphasis is often given to the shape of ceramics and the purpose of our simplistic formulation was to shake this common thought.
Regarding the principle Prof. Read propose: "shared target shape" → "similarity of outcome", we would state it a bit different. We recently studied the potter’s capacity to reproduce familiar vs. unfamiliar types (Gandon et al., under revision). The familiar types were reproduced without any models (only a ‘practical experience’) while the unfamiliar types (cylinder, bowl, sphere, and vase) were reproduced from a concrete 2D model displayed all along the experiment duration. Results showed that the familiar types were reproduced with a significant individual and cultural standardization. On the contrary, unfamiliar types were reproduced with a low standardization. In fact, each potter of the group reproduced unfamiliar types in an individual manner, distinct from the other potters. These inter-individual variations were particularly salient for the two more complex shapes (bowl and vase). Then, even with the same target (i.e. same 2D model), the outcomes were significantly different. This leads us to the conclusion that the similarity of outcomes requires not only a shared target but also a minimum period of practice during which potters of a given group can train for the shaping of new types. Then the principle "shared target shape" → "similarity of outcome" would be verified systematically for familiar shapes and only for simple unfamiliar shapes, as it is the case for the spherical shape tested in the Plos One article.
Finally we would like to underline the fact that the message of the article is the functional equivalence of culturally distinct motor traditions, and not the complex relation between the shape standardization and the cultural origins of their makers. In our view, this latter issue will be considerably enlightened by an analysis of the relation between the hand positions used by the potters and the geometry of the resulting pots.