Cheryl Says She's An Atheist

[koganbot]

Proposal for a social psychology experiment:

We'll use four separate, sizable groups of people, say 75 people in each group. (Not that I know if that amount is any good or not, or if we want our overall pool to be similar socioeconomically. I'm not a statistician.)

Ask each member of Group One:

What arguments would you use to try and persuade an atheist to consider that there might be a God after all?

[It's likely that at least a few people in each group will be atheists, but that's no reason they shouldn't try to answer the question.]

Ask each member of Group Two:

Cheryl tells you she is an atheist. What arguments would you use to try and persuade her there might be a God after all?

We're trying to see if by giving our atheist a name, so a potential personal, individual history, we elicit responses here and there that are different in type from what we generally got in Group One.

Ask two questions of each member of Group Three:

Group Three Question 1: Cheryl says she is an atheist. What arguments would you use to try and persuade her there might be a God after all?

It's important that the subjects complete the first question before seeing the second.

Group Three Question 2: What do you imagine Cheryl's reasons might be for being an atheist?

Then for Group Four, we reverse the order of the two questions, again making sure the subjects finish the first before seeing the second.

Group Four Question 1: Cheryl says she is an atheist. What do you imagine her reasons might be for being an atheist?

Group Four Question 2: What arguments would you use to try and persuade her there might be a God after all?

There are all sorts of ways to tweak these questions, were we to have access to a large population of potential subjects who were willing to respond to our questions. For instance, for a fifth and sixth group we could replace "What do you imagine Cheryl's reasons are for being an atheist?" with "What do you imagine are the various beliefs that make up Cheryl's atheism?"

For the fun of it we could have groups for whom the name of our atheist is Jason, not Cheryl, to see if there are differences between how our subjects go about persuading a woman (albeit an imaginary one) as opposed to a man. Of course, we can also notice the differences between how male and female subjects undertake to persuade, and I suppose we could look at other demographic subsets within our groups — though slicing up our results demographically may result in subsets that are too small to give statistically meaningful results. E.g., if there are eight Asians among a group of seventy-five, those eight are not a big enough sample to generalize from at all about Asian attitudes.

But returning to the main reason for the survey: What I'm expecting to happen is that in Group Two, and Group Three when regarding Question 1, a greater percentage of people will be a bit more uncertain than those in Group One as to what arguments to use; that a greater percentage in Groups Two and Three than in Group One will offer multiple different arguments rather than just one or two, not being certain which arguments will speak to a particular person; and that many people in Group Four will be more uncertain still, a number of them being more flexible and creative and a number of others more puzzled and paralyzed. And I'm predicting that this'll even be true of the atheists among our subjects (atheists in Group Four being more uncertain and more flexible regarding what arguments will be necessary or effective than are the atheists in Groups One, Two, and Three), though presumably atheists in any group will be pretty uncertain anyway about what would be persuasive, given that they've not been persuaded.

Four hypotheses I'm testing:

--First, that people are more certain about what an unspecified member of a group will believe than about what an identified individual might believe, strange as that might seem, and even when the identification is exceedingly minimal, just a name.

--Second, that people have no idea how clueless they are regarding other people's beliefs. What Secretary Of Defense Donald Rumsfeld said about there being "known unknowns" and "unknown unknowns" was self-serving in its context, the potential existence of weapons of mass destruction in Iraq, and their potential export, which became an American justification for invading — I assume (though I don't know this) that a lot shouldn't have been even in the "unknown" category, much less in the "unknown unknown," i.e., the category where we don't even know that we should look for something, much less where. But there certainly are genuinely unknown unknowns (I'd say that discontinuous energy levels were an unknown unknown for physicists in the 1890s, though someone who knows the history of physics, and physics itself, better than I do might know or suspect that I'm wrong), and to some extent the potential fact of them can be prepared for. Budgeted for, anyway. My experience is that, when it comes to other people's ideas, there's a lot of knowable stuff that could easily be in the "known" category or the "known unknown" category but that for most people is in the "unknown unknown" category. They think they're in known territory and simply don't know that there's something there they don't know. (Not that I work for a university or have access to funds to conduct a massive experiment of the sort I'm proposing. Think of this post as being an oblique elaboration on my Dead Lester post, both this post and that one, among others, getting the "mutual incomprehension pact" tag.)

--Third, that when we give people a hint that they don't know something, or remind them of it, at least some of them will respond with more uncertainty and more willingness to imagine alternatives. That's why I'd expect a fairly big difference in the response to Group Three's Question 1 and Group Four's Question 2, even though they're the same question. If you've already been prodded to wonder if Cheryl has a number of different reasons for being an atheist, and whether or not you know all of them, you might then be less certain and more imaginative as to what ideas might persuade her to rethink. I also expect there will be at least some difference between responses to Group Three's Question 2 and Group Four's Question 1, that the people in Group Three, in trying to think of what would persuade Cheryl, and having therefore come up with some instant rudimentary ideas as to what Cheryl's atheist beliefs are and her reasons for them, will be somewhat prone to stick with those reasons when asked for them in Question Two. I'd expect fewer people to respond with "How the fuck would I know?" to Group Three Question 2 than to Group Four Question 1.

--Four, that my reasons for being an atheist aren't even on most people's maps, and that few of my beliefs will show up in the ideas that people ascribe to Cheryl. (I bet not many people in the world would guess that one of my thoughts when reading Nate Silver on opinions converging towards the truth was that he was being insufficiently atheist. I bet he'd be surprised.</Frank being cryptically provocative>)

This experiment would also be something of a fishing expedition: maybe people would come up with reasons for being an atheist and reasons for believing in God that I hadn't thought of. Maybe there would be surprises regarding what people imagine will persuade other people, some unknown unknowns. I'd hope that at least a few people in Group One would ask "Which atheist?" (Give that person a job interview!) Maybe I'm all wet and the vast majority of subjects will feel the full uncertainty of "Cheryl says she's an atheist." But conversely I fear that the question is so daunting that we wouldn't get enough useful responses, that people would give the written equivalent of monosyllables and grunts. (Maybe we'd offer incentives or threats, tell you that you won't get paid unless you write at least four sentences for each question. Or maybe we'll appeal to people's altruism by begging and crying.) Perhaps we'll get angry retorts from polytheists for referring to "God" rather than to "the gods." I hope so.

Comments

[arbitrary-greay.livejournal.com]

The dynamics discussed here are found in the competitive debate circuit. In the CX/Policy division, (the one I have the most experience with and prefer) the affirmative team reads their plan and justifications, the negative team reads a combination of arguments directly targetting the affirmative's first speech and arguments that function as their own independent unit that the affirmative team can dismantle on multiple levels.
(Remaining speeches consist primarily of line-by-line discussion, and then attempts at consolidation and conclusions in the last speeches. The success of which depends on the skill of the debaters, and whether or not they actually figured out how to disengage from the line-by-line, which I did not before leaving Debate.)

The 1st affirmative speech is canned, although perhaps the team, familiar with their opponents' tendencies, might pack in a few preemptive justifications.

Where the dynamics you discuss here come into play is in the 1st negative speech. As various 1st affirmative speeches become familiar to teams during the season, arguments for the negative also become standardized, and the 1st negative speech can also become canned. (there are even some canned negative argument units that rely on the slimmest of connections to the affirmative so that they can be used across seasons and regardless of how different the details of various affirmatives are)

But just as the same affirmative plan can be customized to different intents through different justifications, a better debater will customize the 1st negative speech to the specific justifications of the affirmative. A lesser debater will rely on the same canned arguments for the same plan regardless of the justifications.

And in general, the quality of debates increases moving away from canned blocks of text defending arguments previously put out to being able to intimately deconstruct the text of the opponent. However, it is a rule that new arguments are not allowed in the latter half of the match, (so that you have only have to defend affirmative, negative attack, affirmative counter-attack, and negative counter-counter attack) so the better teams are the ones who are able to produce the more specific attacks from the start, as they will be less likely to need to produce new arguments as counters and risk losing the point to tangents, and can instead focus on strengthening their position in the eyes of the judge. In practice, this must start from knowing the affirmative's evidence as intimately as one's own.

So Group 1 is the least ideal situation, akin to the first debates of the season when affirmatives are yet unknown. One can produce a canned negative 1st speech of generic arguments common to the topic, but odds are that the quality of the overall discussion will be quite low because of instances like yours where the subject can sidestep everything saying "But I agree with you. My justifications follow a different intent than the direction of your arguments." Ships pass in the night.
Or the negative can play framework games instead, attack the "bigger picture" so that all affirmatives apply no matter their details. Like criticizing how the atheist, simply by taking a stance, inherently creates a boundary by which believers are judged. Or disputing their definition of "god" and presenting a counter-definition that ensures the existence of god/s according to the affirmative's arguments.

But anyways, you could get a glimpse of how people would respond to your hypothetical study by observing the trends in how debate teams formulate their 1st negative speeches.

[christophe andersen (from livejournal.com)]

"What arguments would you use to try and persuade her there might be a God after all?"

I think this question is a bit unclear and you would possibly get a variety of types of responses. Parsing this question out, I think it is asking for arguments to persuade an atheist to really be agnostic. I don't think that is what you mean for it to ask, however I could easily be wrong with that assumption. Most people would probably respond to the question by creating an argument to go from atheism to theism. They'd attempt to convince the atheist that their specific version of god does exist. Then there would be people like me that get stuck on the "might exist" part, as the question is written, which again is really an argument for agnosticism and not theism. While both types of arguments would be interesting to study, I imaging it is a lot easier to convince an atheist to be an agnostic rather than a theist.

[koganbot]

The question is clear but doesn't give enough information (it doesn't tell us why Cheryl is an atheist, it doesn't tell us what she thinks being an atheist means, it doesn't tell us what she thinks believing in a God means, and of course it doesn't tell us what sort of arguments persuade her; some people think it's persuasive to tell someone "You're not being open-minded"). You're mostly right in saying the question is asking "How do you persuade her to be an agnostic?" except that "There might be a God after all" is actually a bit stronger than "I don't know if there is or isn't a God," which just says that you can't prove otherwise. (And there are other definitions of "agnosticism" too, that I won't go into.) I'd say, following along with what you wrote, that in responding to questions, people (incl. me) respond to the question that is intended rather than the question that is literally asked. And we're right to do so, if we know what question is intended. But people also tend — this is inescapable, but some people are more prone to this than others — not to notice that they've substituted the intended question for the literal, and may in fact have been mistaken as to the intended question. Ditto, people may forget to consider that what was said may not be the question that the questioner actually intended to ask.

But what I'm specifically after in this experiment is to test the hypothesis that people way underestimate how much information they don't have and how much uncertainty they should have (I might say "in general," though this experiment doesn't test the "general," instead picks an issue where there ought to be very high uncertainty), and to test if, by adding a name and shifting the question order, we can increase the uncertainty and the desire for more information.

Any knowledge I get about actual attitudes and arguments and beliefs regarding God and atheists is gravy. Unfortunately, I doubt this experiment will ever take place (I'm not in position to conduct it), and it probably wouldn't work anyway, the question being too daunting for many of the participants. But maybe if we ran the experiment on people we assume are motivated to answer — college students, churchgoers — we still might get useful answers (a hypothesis being that we get a significantly different response from members of Group Four than Group One, for instance, no matter the social makeup of the subjects; which doesn't mean that, if we use a socially homogeneous bunch of subjects, we don't have to then try the experiment on a bunch that's socially different from the first one). Anyway, this is a thought experiment more than anything else, and an oblique commentary on my world.

[koganbot]

This is interesting. I was on debate team freshman year in high school (this is 44 years ago), didn't do very well. I see the similarity regarding canned arguments versus responsive ones. One thing debate has, though, is a judge, who, if she's any good, will penalize a team for not responding to the other team's ideas (or at least for not giving a good reason for the nonresponse), for attacking straw men, etc. Also, it seemed to me at the time that, among other things, debaters had to either have a clear idea of the weaknesses in their position and know how to cover them up, or have no idea of the weaknesses in their position and for that reason be good at covering them up (the cynic in me says that the latter do even better than the former).

One difference in my experiment is that the subjects have to think about persuading the other party (rather than impressing a judge). Another is to see what increases uncertainty, and therefore (one hopes) produces a more flexible and creative response, and a thirst for knowledge.

Of course, uncertainty itself can be a debater's tool. One reason I tend to find Paul Krugman and Nate Silver persuasive is that they'll sometimes refer to alternate hypotheses or readings of the data, rather than just pushing for their favored reading. This, interestingly enough, tends to make me more prone to believe their favored reading, since they've seemed to have questioned it and tested it themselves. It also makes them more interesting to read. But in fact, I've not seen Silver's data and I'm far from mastering Krugman's ideas; but I'm prone to believe those two over their opponents anyway, not just 'cause I'm registered to Silver's and Krugman's political party but because of the way they play the uncertainty card.

While there isn't a direct analogy between debate and my experiment, I'd bet, not cynically, that people who become good at formal debate (as opposed to TV polemics) would also become good at responding to other people's actual ideas, recognizing when their own canned responses don't take account of those ideas, recognizing weak points in their own arguments, recognizing when there might be alternative hypotheses, and knowing when they could use more information, where there might be known unknowns and unknown unknowns.

[arbitrary-greay.livejournal.com]

You're exactly right about the differences. I would use drastically different tactics in debating vs. persuasion, namely that I would play all sorts of unfair philosophy games and blow up inconsequential flaws into the judging points. In discussion, while I still love many of the concepts used in debate, I do try to avoid getting bogged down in semantics as the main point. (Semantics always make fun friendly tangential discussions) I also tend to make lots and lots of concessions that I wouldn't in competition for the sake of deepening discussion instead of butting heads over the same points over and over.

On the other hand, there is an aspect to persuading the opposing team, even though the judge is subject in this case, because there's no stronger reason for a team to win than forcing their opponents to concede something. (Although again, in a competitive mode I'd be aiming for something very small which I'd blow up to importance later) Or the next best thing, re-routing your own opponents' argument against them, as "Your arguments against the existence of god are false" isn't nearly as convincing as "Your arguments against the existence of god actually prove the existence of god."

But the differences are why the parallel to your experiment lies strictly in the first negative speech, before the competitive part of the activity exerts its influence. Ignoring teams being familiar with the others' preferred tactics and counter-arguments, (or, the subject's gender/ethnicity/social and economic status/history) the only thing that the negative theoretically knows going in is that the affirmative is the affirmative. (or, the subject is an atheist)
How the various groups in the experiment increase in the amount of information they have going in is akin to how debate teams increase in their knowledge of opposing teams and arguments over the course of the season. At the beginning of it, however, all they know are the arguments they have prepared going in, which is Group 1.

Krugman and Silver: I wouldn't call that uncertainty, but pre-emption. They deal with the common potential counter-arguments up front, which in competition would waste valuable time as both sides go through the motions making standard counter-arguments in the next speech. (And time-wasting is a tactic) This produces a better discussion as trivial-but-common counter-arguments are discarded and the conversation can move to unexplored ground.

[koganbot]

I probably didn't present Krugman and Silver all that well, but what they do that impresses me is not that they pre-empt standard counterarguments that they know are wrong (though they certainly do that, and it's good that they do), they also present counterarguments and alternative hypotheses that they believe have something to them. Which means that when they're very certain and aren't presenting counterarguments, I'm also likely to go with them (even when I don't understand what they're saying), since they're the sort to present uncertainty when it's warranted.

In any event, here's Krugman recently on an idea that he and some other people are playing with (technological advances making workers superfluous in some areas, hence an increase in inequality), and it's potentially very important, he says, but the idea is only in the formulation stage.

http://krugman.blogs.nytimes.com/2012/12/08/rise-of-the-robots

Additional factors or explanations cited:

http://krugman.blogs.nytimes.com/2012/12/09/technology-or-monopoly-power

More detailed explanation, potential scenarios:

http://krugman.blogs.nytimes.com/2012/12/10/technology-and-wages-the-analytics-wonkish

What's happening right now is that we are seeing a significant shift of income away from labor at the same time that we’re seeing new technologies that look, on a cursory overview, as if they’re capital-biased. So we could be looking at my technology B story above.

Supporting evidence:

http://krugman.blogs.nytimes.com/2012/12/11/human-versus-physical-capital

[skyecaptain.livejournal.com]

Currently reading Us Against Them: The Ethnocentric Foundations of American Politics. This has been a good foundational reading in in- and out-group identification, which seems to be at least part of your interest here. The in- or (more likely) out-group would be "atheists," depending on the respondent's position, and it's likely that people in the in-group AND out-group underestimate how much they know about "Cheryl," even if they'd be quick to define "atheists" according to their at-hand beliefs.

From the book:

[M]embership is not sufficient to establish an in-group, just as the absence of membership is not sufficient to establish an out-group. What is required is psychological striving: attraction and identification in the case of in-groups; condescension and opposition in the case of out-groups.

The book persuasively argues that ethnocentrism, though a universal feature of human psychology, is more apparent or less apparent in particular groups and individuals, and is not always related to cogent "group interests." (That is, there aren't always strong reasons why individuals are or are not more ethnocentric than others; rather they have general predisposition to ethnocentrism in general, as opposed to, e.g., prejudices against particular out-groups.)

This might be relevant to a few parts of your thought experiment. Firstly, there is likely a correlation between the strength of your belief that atheists do in fact constitute an in- or out-group and the certainty with which you ascribe beliefs in the general category ("an atheist") question. This may or may not have an effect on how certain or uncertain people are in regard to Cheryl, though I would bet the strength of out-group identification will be reduced, though probably not eliminated (or even that much reduced).

Second, as long as out-group identification is a core issue, reasoning will be secondary. By introducing Cheryl, you're diminishing the ease with which someone can fall back on their out-group identification and essentially taking out -- or at least diminishing the power of -- the battery that fuels their certainty, which may or may not have any relationship to reason. What's not clear to me is whether or not such an intervention actually in the end decreases that out-group identification. I imagine it does. But I also imagine that if you followed up by returning to the out-group in general there would be a high level of cognitive dissonance as people used the same (poor) reasoning to describe the out-group even immediately after using (better) reasoning to say perhaps the opposite, or something quite different about an individual.

[skyecaptain.livejournal.com]

Re: Dead Lester, it seems to me that rock crit is in its own way uniquely ethnocentric -- that is, particularly prone to forming with and against groups. It's a reason that lots of people become passionate about music in the first place. When I "got into music," it was explicitly because I wanted to define myself with a particular group -- the "music kids" -- and my desire to be part of that group would have, at that point in my emerging understanding of music -- diminished the power good arguments counter to their positions, no matter how absurd their positions might have been. That was my psychological striving, and to the extent that I was striving, I would have been far less likely to challenge any beliefs that happened to jibe with the tribe.

It seems like rock critics -- including me -- are often members of a tribe pretending to do anthropology on other tribes. This is (in part) what makes them "lousy anthropologists" -- they're reconfirming the boundaries of their in-groups rather than (somewhat dispassionately) exploring the various dynamics at work. If their experience with music and social relationships was anything like mine, they don't have a research question; they have instead inchoate needs to confirm their participation in a particular group. That's part of what led me to the music convo in the first place, and my somewhat radical departure from in-group-ism (or at least a more selective view of what counts as "in-group") was largely due to a sense that I was being conned, that this group wasn't right for me. ("No group is right for me" is a pretty strongly anti-ethnocentric position, I would think. It means that my blindnesses, though often apparent, are not always predictable, and that I'm fairly likely to accept a good counter-argument. I think I'm also prone to mentorism, striving to find a particular model rather than an accepting group.)

Tip of the iceberg -- quick scan on Google Scholar led to this highly-cited study from the 90s -- don't know enough about social psychology stuff to dig much further (for now). The abstract:

This study investigated the processing consequences of receiving non-membership-relevant persuasive messages from in-group or out-group members. Students were given two-sided messages ostensibly from an in-group or out-group source. The position advocated in the message was announced either before or after message arguments were presented, and position-consistent arguments were either strong or weak. In-group messages were more likely to receive content-focused processing (as indicated by lager processing times and differential persuasion to strong and weak arguments) when position advocacy followed rather than pre ceded message presentation. Prior knowledge of the in-group position produced acceptance of the in-group position regardless of message quality, particularly of the counter attitudinal message. Out-group appeals produced almost no attitude change, even with strong arguments.

The subjects here were likely university students (as are many social psychology experiments). Knowing the position going in led to less engagement in the content of the message, though I don't know how much less from the abstract. (This might be a Dead Lester problem, too, that we're usually fairly sure of the position going in.) And out-group persuasive appeals -- even high-quality ones -- usually do little to convince an in-group of an alternative position. Though you would likely find more frustration and confusion in using "Cheryl" instead of "an atheist," I'm not sure what one would do to actually counteract the assumptions that your participants will likely bring to the table. Recognizing one's ignorance in the moment is one thing, but doing something about it is another (and beyond the scope of your social experiment, but not beyond the scope of Dead Lester).

[koganbot]

it's likely that people in the in-group AND out-group underestimate how much they know about "Cheryl," even if they'd be quick to define "atheists" according to their at-hand beliefs

Assuming you didn't mistype and mean "overestimate" rather than "underestimate," this is the opposite of what I'd expect. That is, this says that people think they don't know as much about Cheryl as they actually do, and that they therefore have more uncertainty than they should. If this is what Kinder is saying, what does he base it on?

Or is he saying that they often go to their immediate stereotype assumptions but actually know more than their initial response indicates?

My experiment doesn't test these directly; it just tries to see if introducing a name and switching the order of the questions would induce people to have (even?) more uncertainty.

It's certainly not my experience that people (atheist or nonatheist) know more about my beliefs than they think they do, though it's possible that they know more about the average atheist than they think they do (I'm assuming that I'm not an average atheist and that therefore, by probability, a randomly chosen, imaginary atheist such as Cheryl is likely to be closer to average than is someone like me).

[skyecaptain.livejournal.com]

I did mistype -- meant to say "underestimate how much they DON'T know"! Woops!

[skyecaptain.livejournal.com]

Haven't gotten far enough into the book yet, but generally Kinder and Kam so far suggest that out-group identification (and vehemence of that identification) is essentially unrelated to information about the group. One can have a high level of knowledge of an out-group and still be generally condescending and hostile to them, or one can use stereotypes and misinformation to bolster beliefs about large groups. They may go on to clarify to what degree knowledge and attitudes are related or unrelated but I'm not sure yet.

[koganbot]

Whew! That's a relief.

[skyecaptain.livejournal.com]

If I had to guess, though, I would assume that "ignorant out-group identification" is far more common than "knowledgable out-group identification." I'm interested in the second kind, though -- it's what I identify in a lot of progressive groups I'm associated with -- an unwillingness to even read the opposing side because of what they (legitimately) know of the side in question. One thing that attracts me to, e.g., Matt Yglesias (along with Krugman and Silver) is (as you say) his ability to articulate the other side's position and act in good faith in providing his own analyses. It leads him to assuming many very progressive and also very traditionally conservative positions (his stance on deregulation of zoning policy, the subject of his ebook "The Rent Is Too Damn High," is actually a strongly free market argument, despite it not being on the radar screen of conservatives or liberals, who tend to resort to the more prosaic forms of ethnocentrism to be found in Not In My Backyard-ism).

[koganbot]

I remember reading* that audience members for "point-counterpoint"–type debates on TV don't learn anything about the ideas they go in opposing, since they basically just listen to their own side and dismiss the other, without making the effort to comprehend the other. (But what about uncommitted viewers?)

*Yes, this is not very good sourcing or fact-checking.

[skyecaptain.livejournal.com]

Another interesting point in the opening chapters of Us Against Them is the aassertion that ethnocentrism can function as both strong in-group and strong out-group identification, but doesn't require one or the other to have the same effect. That is, you can identify lots of out-groups without having an "in-group," but you can also identify with an in-group that doesn't necessarily correlate to some deep-seated hatred of others. "[E]thnocentrism need not be interpreted as a dark and irrational expression of repressed hostilities and primeval fears. Ethnocentrism is a commonplace consequence of the human striving for self-regard and personal security."

[skyecaptain.livejournal.com]

Kahneman describes some of the cognitive dissonance that can occur when you personalize a generalizable quality, too -- when students in his psych class watch personal interviews with someone whom they know are statistically likely to act in X fashion, they almost without fail make that person the "exception" to the statistical reality. Out-group identification is its own kind of generalizable reality (one that is not always accurate) from which people can exceptionalize an individual without affecting the way they think about the group. But IIRC Kahneman also mentions a few strategies for diminishing this effect; don't have the book on hand to double-check.

[koganbot]

I have to break from this discussion for about an hour, but one thing I'm not clear if I'm getting right:

The phrases "in-group identification" and "out-group identification" — does the former mean that you identify something or someone as belonging to your group (and identify yourself as belong to that entity's or person's group?), and the latter that you identify something or someone as belonging to a group you're at odds with or outside of (and identify yourself as outside of and at odds with that group)?

The trouble is that when I see the phrase "out-group identification" my immediate impulse isn't to think, "I'm identifying them as part of a group I'm outside of and at odds with," but rather, "I identify myself as belonging to this particular out-group." The phrase "group-identification" usually refers to what the subject identifies with, not to where the subject locates someone or something else.

[skyecaptain.livejournal.com]

I should be clearer in when I use "identification." Generally in-group is used to mean "I identify as a part of this group." You identify against an out-group. "We need to keep X out of our community" would be a typical example.

You can identify with a group that others consider an out-group, but you probably wouldn't call that an "out group" unless you were referring to the position of some other group of people "Jewish" is historically seen as an "out group" from some perspectives; my wife considers it an in group, but though I might be considered as part of the "out group" by others -- meaning if I were born in Germany in the 1920s like my grandfather was I'd be identified as part of the group -- I don't identify as Jewish.

[skyecaptain.livejournal.com]

I think the value you brought to the rockism debate was in pointing out that there didn't seem to be any coherent "in-group" there -- even though anti-rockists defined against "the rockist," no one could really come forward to say "I'm a rockist." (And further, if "a rockist" means values X, Y, and Z, then maybe Frank Kogan is a rockist, and what's wrong with that?) Whereas "punk" is strongest as an in-group identification even though it could also be used as something to define against. ("Keep the punks out of our community" makes sense in a way that "keep the rockists out of our community" never quite did.)

[koganbot]

A lot of people do identify themselves as belonging to outgroups. I think freaks think of themselves as an outgroup, skaters think of themselves as an outgroup, punks think of themselves as an outgroup, indie kids think of themselves as an outgroup, alternative thinks of itself as an outgroup, "real Christians" think of themselves as an outgroup (a besieged one), etc. etc. etc.

But more to the point, when you say "Frank's out-group identification," you're sending a signal that you're talking about what out group Frank identifies with, not what out group Frank slots other people into.

[skyecaptain.livejournal.com]

I think that "identifying as an outgroup" is an interesting phenomenon, especially if it doesn't correlate with "identifying with an in-group." Though I'm not sure if any of your examples genuinely don't have corresponding "in-groups." It's certainly pertinent to the rockcrit world, and maybe even liberal pockets, where identification of bad others (and the repression of My Good Ideas) don't always correspond to a clearly defined "in group."

But I'm not sure if this particular book addresses this facet of identification -- they're more interested in general statements disparaging or marginalizing others as "out groups" while identifying as "in groups." But I could be wrong.

(When you identify with a marginalized group, though, you're still generally identifying in-group, even though you can see that other people slot you into the out group category. "Out grouping" is an act of differentiation and exclusion.)

[koganbot]

I'm thinking as an editor here, in case you end up writing about this elsewhere: if you're addressing an American audience and you say "Person X's in-group identification," the phrase is going to suggest an in group that Person X identifies with, and if you say "Person X's out-group identification," the phrase is going to suggest an out group Person X identifies with, not the people or characteristics that Person X assign to an out group to which Person X does not belong. And if you're speaking broadly about in groups and out groups, the reader is going to think of minorities and outcasts etc. when you say "out group," even when what you mean is something different.

This doesn't mean that you shouldn't talk about an "in group of punks," or an "in group of transvestites," or whatever. Just don't ever use the phrase "out-group identification" for what these particular in groups are defining themselves against. Don't use that phrase at all. Instead, say something like, "characteristics that the in group of punks assign to groups they define themselves against" or "people the in group of punks assign to groups the punks define themselves against" or whatever particular point you're trying to make at the moment.

[koganbot]

I think I brought more value to the "rockism" debate [scare quotes] than just that.* In fact, I don't altogether agree with what you're saying here. For one thing, I don't think anyone ever claimed that "rockists" constituted a group. I think you're getting too caught up in applying the dynamics that Kinder & Kam describe. For instance, I doubt that atheists think of atheists as constituting a group, in group, out group or otherwise, and they certainly wouldn't think of nonatheists as constituting a group. And maybe some Christians are daft enough to think that atheists constitute a group, but I'm sure a lot of Christians don't, even a lot of Christians who feel besieged by secularism. Which doesn't stop people from projecting views onto atheists that atheists don't hold, or applying stereotypes to whole swaths of people.

If there's an in group at an office, it doesn't follow that there are necessarily out groups, just outsiders. Or if there was, let's say, an in group of punks in London in early 1976, there wasn't necessarily a set of groups they defined themselves against. There may have been ("hippies," metal heads, pub rockers), but there didn't have to be. And the punks probably felt themselves way outside the mainstream, but I doubt they thought of "the mainstream" as a group. They wouldn't have identified with the Greater London Commerce Association (if there ever was such an organization; I made up the name), but if someone introduced himself to Johnny Rotten as representing the Greater London Commerce Association, Johnny would likely have felt a social chasm between them — the group name identifying the fellow as mainstream, even though Johnny'd never heard of and had no previous opinion on the group. Defining yourself against others isn't the same thing as defining yourself against other particular groups. (Not going into it here, but that's why the actual functioning of "social class" is hard to get a grip on, since people tend not to congregate and aggregate in what we normally think of as social classes.)

*Or will bring value if anyone other than you and me and Mark actually decides to think about what I said. I'd say more to the point is that the antirockists projected their own authenticity impulses onto the supposed "rockist" but in stupid form, and then skewered the "rockist" for his stupidity, thereby achieving a cheap victory over an imaginary foe while leaving unexamined all the actual social and class issues that swirled about those authenticity impulses. (Not that everyone who decried "rockism" was doing this.) Anyhow, none of this means that people saw "rockists" as constituting a group.

What I'm saying here is all pretty tangential to the interesting stuff you wrote upthread; I'm thinking the Kinder-Kam dynamics might apply to vaguely conceived others, not just to particular, identifiable groups (or to identified groups, anyway, even if the groups aren't as coherent as those defining themselves against them think).

[koganbot]

Interesting that they have to use the word "ethnocentrism" when actual ethnicities rarely seem to be at issue.

[skyecaptain.livejournal.com]

Yeah, a few cases in this thread of typing faster than thinking (which of course a good editor is supposed to catch...but I probably should rely on myself as an editor more often). FWIW the authors never use the phrase "out-group identification," since in their definition of it (so far) that would be an oxymoron. In-group identification, conversely, is complicated, since often one does define themselves as an in-group in the abstract but not the specific (that is, there are atheists, and a category of people called atheists, but I don't identify with a cogent group that might meet and become a National League of Atheists together).

This is in part where the "more or less ethnocentric" stuff probably comes in -- and of course I'll have more to say about it when I've actually finished the book (what a concept!). I would guess that given the authors' definition and measures of ethnocentrism, most atheists and agnostics would be generally more tolerant of lots of different groups, hence "less ethnocentric" and likely also less likely to identify strongly with or against a group. But again these are just my guesses, not what the book says. (I'm using this comment thread somewhat selfishly to get thoughts that have been flitting around for a week out somewhere.)

[skyecaptain.livejournal.com]

Ha, I am often guilty of projecting whatever I'm reading at the moment onto EVERYTHING. Guilty here, too. (There's more to be said about the in-group/out-group stuff and rockism/anti-rockism -- the trick is in moving away from the "concrete group" thing and focusing more on that psychological chasm between self and other, which is closer to how out-group hostility actually manifests itself -- but I need to read more and stew on it for a bit.)