Rogue Thoughts

I found these old reviews from when I submitted the article to Religion. The peer reviews were generally very positive, but the editors remarks were not that positive. See for your self:

Reviewers’ comments:




Reviewer #1:   

Against anthropological orthodoxy, which locates the responsibility for the credibility of divination practices in characteristics of diviners, this paper advances what might be called the “cognitive repair” hypothesis.  That hypothesis holds that it is the mind’s automatic attempts to resolve the puzzle that the ritual actions present that is the source of divination practices’ credibility with clients.  Rituals displace the intentional component of action, provoking a search for an underlying agent.  This leads to inferences about unobvious agents, who are, therefore, attributed counter-intuitive properties.  Those properties include access to knowledge that these agents reveal via their influence on the patterns diviners interpret.  Thus, the author concludes, divination practices gain their credibility (primarily?) from clients’ confidence in the influence of these unusual agents on the divinations’ outcomes.  The paper is mostly about two empirical
studies.  The first looks at subjects’ perceptions of various actions by a diviner and finds, in a design that disguises the study’s goal, that subjects are significantly more likely to find the resulting advice credible in ritual actions than in alternative actions in which the diviner intentionally established the pattern. Although agreeing with Boyer that ritual credibility turns on the displacement of intention, the author maintains that it turns, specifically, on the status of the posited agent who is behind the whole transaction.  The second study tests this subsidiary hypothesis. 
        This paper is clear (for the most part), well-organized, and direct.  It demonstrates that hypotheses in this area have testable consequences and offers a plausible program of research for testing them, in fact.  It also exhibits a level of sophistication with psychological experimentation, experimental design, and statistical analysis that is refreshing in the literature on divination, ritual, religion, etc.  Attention to the following items (which are in no particular order) would improve the paper.

1. In the paragraph at the top of p. 4 is insufficiently restrictive to characterize divination (other than by using the language of divination throughout).  Why, for example, would science and scientists not also satisfy the conditions the author provides?
2.  The introductory account (near the top of p. leading to the discussion of the first experiment is inaccurate.  The dependent measure is not “the likelihood that the main character acted on the information” but rather subjects’ ratings of that likelihood.
3.  To avoid putting off some readers for the wrong reasons, alternatives to the language of “primitive” and “modern” people on p. 9 might be wise.  (The point, after all, is not about the people but about their cultural settings.)

4.  Having two independent raters and doing the Cohen Kappa test for their agreement would be more convincing than checking the agreement of one rater with the experimenter. 

5.  In the middle of p. 10, the author introduces the second measure about subjects’ ratings of the likelihood that the client found the advice worthwhile, however, the author does not discuss (here) what this measure safeguards against.  It is not obvious.

6.  Expand a bit on the relevant material at the very top of p. 10, so that in the section on “Design” that follows it will be more clear what “(Sets)” is.

7.  The large paragraph on p. 11 is unclear on multiple fronts.  Say explicitly what ACC and AD are.  Say which rating is higher in the Kurabi story.  In what sense is that rating “in this set” (as opposed to “with this set”)?  Clarify what is referred to by “Action type” in the final sentence.  (There is ambiguity here in the use of the term “Action,” given all that the paragraph discusses.)  Is the preposition in the final sentence of this paragraph correct?  Should it be “of” Action type, instead of “on” Action type? 
8. The opening sentence of the second paragraph on p. 14 is oddly framed.  It is “possible” to attribute the effect to universal processes, but that they are “not culturally variable” is not something that the experimental evidence provided here supports.  At a minimum, that would require carrying out the study cross-culturally.   

9.  The first rationale on p. 15 for experiment two does not seem to the point.  Experiment 2 tests whether or not the status of a putative associated agent influences subjects’ judgments about others’ assessments of the credibility of a divination, per the account of the experiment provided in the next paragraph. 

10.  The author should state clearly that the results of experiment two fail to distinguish between two possible measures of a god’s prestige, viz., the frequency with which sacrifices are made to the god as opposed to the value of the sacrifices made to the god.
11.  The first two sentences of the middle paragraph on p. 18 are puzzling.  How does the comment about non-believers’ probable preferences follow?  Indeed, it is seems inconsistent with the comment about the preference of “people in general” in the first sentence.

12.  The discussion in the penultimate paragraph on p. 19 would be enhanced by including an example from the ethnographic literature.




Reviewer #2:

This is a fine contribution to the cognitive science of religion, interesting, original, and well written. Some typos still remain, though (e.g. “Litteratur;” Boyer & Ramble is listed twice). Methodology is quite alright. However, is the author(s) wish to make the contribution still stronger, the concepts of “agency” and “counterintuitiveness” could be explained in more detail. Barrett’s (2000, 2004) idea of HADD thus might be included. You do not make it clear enough what you actually mean by counter-intuitive agents. Some references with this in mind:

Boyer, Pascal, & Harold Clark Barrett. (2005). Domain-specificity and intuitive ontology. In David Buss (Ed.), The Handbook of evolutionary psychology, 96-118. Hoboken, NJ: John Wiley. [about agency detection]

Blakemore, S.-J., P. Boyer, M. Pachot-Clouard, Andrew Meltzoff, C. Segerbarth, & J. Decety. (2003). The detection of contingency and animacy from simple animations in the human brain. Cerebral Cortex 13, 837-44. [about agency detection]

Guthrie, Stewart (E.). (1993). Faces in the clouds. New York: Oxford University Press. [A classic!]

Pyysiäinen, Ilkka, Marjaana Lindeman, & Timo Honkela. (2003). Counterintuitiveness as the hallmark of religiosity. Religion 33(4), 341-55.

Saler, Benson, & Charles A. Ziegler. (2006). Atheism and the apotheosis of agency. Temenos 42(2), 7-41.

Wegner, Daniel M. (2002). The illusion of conscious will. Cambridge, MA: MIT Press. [Includes discussion of Ouija]







Reviewer #3:

1.the article is well written and well researched, although covering only a small area oof research the limitations of which the author, it seems, is not aware of.
2. the research findings represent what Danish people who do not practise divination, think about divination. One suspects that practitioners of divination as, for instance, in Africa, act on different principles; and one expects that the author’s hypotheses will be falsified when exposed to research based on serious fieldwork outside European culture.
3. the article couls be improved considerably by reflecting on the methodological problem indicated above (of generalizing the results from the löimited research area) and by taking into consideration research reports on the practice of divination as, for instance, in Africa (saee, for instance, U.H.Danfulani’s dissertation “Pebbles and Deities” (1995)or his contribution to the volume edited by Gerrie ter Haar: “Imagining Evil”, 2007).  

2.

Editors comments

I quite liked your paper’s extended and nuanced engagement with ritual theory in light of a fascinating set of empirical material. However, some theoretical and methodological questions still trouble me. In the end, these concerns are sufficient to lead me to decline to publish this paper in RELIGION. My co-editor has read the paper, the reviews and my comments below, and he concurs. I will comment on your paper in order both to do you the justice of pinpointing my difficulties and to offer you some thoughts regarding alternative approaches that might assist you in revising what is, my critique notwithstanding, an interesting and potentially quite valuable paper.
I go through it page-wise 
We begin with what might appear as a detail: P. 3: “Consequently the question arises: why does the Ouija board seem to take on a personality when obviously there is none?” Actually, in the anecdote you quote on p. 2 the board does not take on a personality at all, but “the Ouijas written words seemed to take on a personality”. It is not the board, but the words! This is a significant difference, one that has important implications for your analysis.
Fundamentally, I am not persuaded by the displacement of intention theory of ritual. Nor do I find the ritual mode of action to be counter-intuitive: it can only be perceived as such on the basis of a very narrow definition of ordinary action (whatever that may be) like ‘Peter eating an apple.’ My general scepticism with this theory aside, it fails to convince me as you present it. You say “that the immediate goal cannot be referred to the beliefs and desires of the agent.” This reflects the observer’s position (as indicated by “is readily inferable”). So, the reference is actually an inference from the observer’s point of view, it seems: an observer is able to infer that Peter’s “goal is to eat because he is hungry (desire), and that he thinks that eating the apple will relieve the hunger (belief).” Now, Peter crosses himself and Paul will not be able to infer his goal, desire, and belief. However, this distinction between transparent and opaque inference does not correspond to that between ordinary and ritual action.  There are many types of ordinary actions where Paul will not be able to draw the same sort of inferences that are possible in the case of Peter’s eating an apple: Say Peter reads a book: Paul will not know his goal (Entertainment? Education? Planning a trip? Reading a game manual? Sexual arousal?), nor his desires, nor his beliefs. If this does not work, then repairing the “deficiency” does not apply either.
I would think that what we have is a case of deference or deferral, rather than the speculative mechanism that you reconstruct. Your experiment actually appears to measure the former. Let’s take the Kurabi story. To begin with, even where the diviner takes the pebbles in his right hand, beats his hand etc., this is not unintentional as you take it to be. It strikes me as very much an intentional act (the diviner certainly acts intentionally!). However, the intentionally executed action is archetypical as H&L would have it (or not encoded by himself as Rappaport would have it), and the resulting effect of is beyond his control—hence it is deferred, and it seems to me that it is the intentional deference in a deferred mode (= not encoded by oneself) plus the expectation of an intentional result that requires the assumption of an agent. So, my theoretical disagreement here has a methodological counterpart, since I don’t believe that your experiments measure quite what you take them to. Or at least, if I have understood you correctly.
(A note on the dendrologist example: I would have expected Eva to be female, but you write “he”.)
Let’s take a look at experiment 2. Here also I have a couple of problems. To begin with, I think your prestige-strategy is problematic, because, as far as I know, in divination it is not necessarily the most prestigious agents who are invoked, but often less-prestigious ones.
As to the special result of your experiment, I think the fact that Para is given “dried bread” is a deformation compared to the other two conditions where the content of the respective meals is not specified. That may, to some extent, account for the greater difference between the Low and Medium prestige conditions than that between the Medium and High.
I am a bit confused about your criticism of Boyer. Discussing your finding you write: “If ascription of divine agency is secondary to divination techniques as Boyer thinks, we would expect no difference in preference of diviner. If on the other hand, the representation of a counter-intuitive agent is central to divination, we should expect the prestige-bias to create a preference for the diviners communicating with gods of high prestige.” (18-19) From this summary and from the way you initially frame your criticism of Boyer on p. 16, it seems that your theory focuses on counter-intuitive agency whereas Boyer focuses on divination techniques (or ritual form). However, this is not at all what you are testing in your experiment, for the experiment tests the impact of “prestige”, i.e. a specific representation of counterintuitive agency. In order to more clearly investigate techniques, the experiment would need to be devised in such a manner that technique would be held to work even without agency. As it stands now, if may well refine a point in your own theory (but see above); however, it does not refute Boyer.
Finally, your example from Cicero puzzles me. I am wondering whether your interpretation is supported by this source, or whether it is made to fit into your scheme. As I read it, admittedly naively, the anecdote shows that the people were dissatisfied because they felt fooled, and because the rules of the game were fooled around with; there was no deferral, as I would have it; the technique did not allow the situation to speak for itself, as Boyer would have it. I see no evidence for “failed displacement of intention” here.
In sum, if I were not the editor of the journal but were rather invited to publish a comment on your paper, I would do so along the lines sketched above. Being the editor, however, I feel that your paper is sufficiently problematic that I cannot publish it. However, if I have misunderstood you completely, or should you feel that you could reframe your argument so as to make a more compelling case (or if I convinced you to modify your theory), please feel free to submit a new version. Your overall approach to the empirical engagement with ritual theory is work of great value, and work that RELIGION is definitely interested in publishing. My concern is with the details, primarily that the fit between experimental design and hypothesis is not tight enough to ground your argument effectively.
I understand that this is a disappointing decision. I hope that this process will have contributed to your work, helping you to refine your theory. In all sincerity, I thank you for you generosity and patience in allowing RELIGION to consider publishing this work.

I could post my replies if anyone found them interesting, but the article in it’s present form reflects these replies.