The ideological academic bubble functions a lot like the Google search bubble, where any new information is pre-filtered according to previous searches and preferences. The result is a sort of echo-chamber where only messages that support a particular viewpoint make it past the filter, creating a feedback loop, where nothing especially new or challenging can get through.
It is perhaps too easy to dismiss this ideological academic filtering as yet another incarnation of ‘ivory tower’ criticism that is often directed at universities. What makes this filtering different, however, is that it has the real potential to cut students off from future prospects, goals and careers, thereby narrowing their horizons, rather than expanding them. Once the filtering becomes especially strong—as it has become in the humanities and much of the social sciences—it can cut students off from actual reality, producing out-of-touch graduates with notions and attitudes shaped by ideologies ‘backed-up’ by cherry-picked data/examples and which seem to explain everything. As I’ve suggested here, such graduates are vulnerable to the Dunning-Kruger effect.
And this filtering is not just harmful to the students; an educational system that has lost touch with reality plays into the hands of budget-conscious administrators and politicians who may see defunding certain departments, faculties and perhaps even entire universities as a convenient answer to the mounting budgetary crises caused by aging baby-boomers.
But, it is easy to be negative. What can be done to address academic ideological filtering?
As I currently see it, popping the academic ideological bubble involves explicitly addressing the origins of the filtering, which owe much to the ‘split’ between the humanities and the sciences, which in turn can be said to break down along the subjective/objective distinction. The humanities, as well as certain strands of the social and human sciences, tend to enforce the traditional subjective/objective split by: 1) invoking ‘phenomenology,’ a philosophico-theoretical construct that privileges subjective experience over objective reality; 2) relying on the theory of the blank slate, which holds that humans are born as ‘blank slates,’ and that all their differences—sexual, racial, and so on—are the result of social and cultural factors and pressures, rather than anything innate; and 3) employing symptomatic reading, which polices cultural artifacts, science, institutions, etc., looking for ‘symptoms’ of ideological bias. And, when these three things work in combination, they can be very dangerous because they attempt to rob the ground out from under any appeal to objective reality and truth.
So, removing the filters that give rise to the academic ideological bubble would involve dislocating phenomenology, ‘blank-slatism’ and symptomatic reading. And by far the best way to loosen the grip of this ideology on the humanities is to turn back to objective reality and the tools that have served humanity very well for the last couple of hundred years: the data, tools and methods of the sciences. In short, there must be an attempt made to help those in the humanities and the social sciences become more literate in science. The point, as I’ve argued in Posthumanism: A Guide for the Perplexed, is to have the humanities and the sciences not just talk to each other more, but to have meaningful conversations. That said, I’ll add that the sciences have more to teach the humanities than vice versa: the sciences are not so vulnerable to being hamstrung by subjectivity and ideology simply because they are driven by a desire for minimizing subjective bias and thus learning the truth about objective reality. Science and technology studies, of the sort generally found in humanities and social science programs, don’t offer anything much beyond criticizing science and caricaturing the scientific method as tools of oppression; as such, they are worse than useless. Scientists themselves have been by far the best and most effective critics of science because they actually understand how science works; I am recommending we in the humanities listen carefully to them.
I’d cite here the work of people like Andy Clark, David McFarland, Alan Liu and Franco Moretti as good examples of the type of cross-disciplinary mixing I’d really like to see more of.
The following scenario—in part inspired by the lapse of thinking and moral panic triggered by the now infamous ‘Google Memo’—will hopefully illustrate how even a very basic science literacy—in this case statistics—would be of benefit to those in the humanities and certain strands of the social sciences.
Let’s imagine that person A thinks that person B is less suited to a job that requires mathematical reasoning than person C, on the grounds that person B is female, and person C male. Let’s also imagine that person A cites statistical data to back-up their decision, which shows that while women are typically better at calculation, men are typically better than women at mathematical reasoning.
(The graphs used here are not the actual graphs from the studies that show these trends; I’ve taken them from here and here, and I’m using them simply to illustrate several points about interpreting statistics correctly.)
However, person A’s simply pointing to such a graph would not be sufficient evidence for saying that person B is less suited to the job in question than person C; this is where a more thorough knowledge of what the data is actually saying is required. The graph does indeed show that men (purple curve) tend to be better than women (green graph) at mathematical reasoning; however, it is also easy to see that there is a huge overlap of men and women. In other words, the majority of men and women are pretty evenly matched in terms of mathematical reasoning, and this pattern holds for most measurements of the abilities of the sexes. Indeed, it is this overlap that shows how person A’s snap judgement about person B’s ability based on the data is actually more likely to be wrong; person A has committed the ecological fallacy, an error illustrated neatly by the image at the top of this post. So, a more thorough knowledge of the statistical data can actually help a biased individual overcome that bias. This also means that the only way of truly determining which candidate has more merit is to treat both as individuals, and test each on an individual scale.
At the same time, the graph does show that there is a difference between men and women, which should not be ignored: looking more closely at the tails of the curves shows us that men are more likely to be found at the highest levels of ability as well as at the lowest levels of ability, a pattern which tends to show up in most measurements of the sexes. This is because, as Steven Pinker shows, men tend to have more variability than women, a situation he bluntly sums up as ‘more prodigies, more idiots.’ This, of course, is absolutely not to say that women can’t be idiots or prodigies: indeed, the tails of the green curve show very clearly that they can be. It’s just that there will be less of them: this is why we might reasonably expect men tend to be ‘over-represented’ in the categories of ‘prodigy’ and ‘idiot.’
As the above scenario shows, an individual with access to solid data can be wrong about things. At the same time, an individual without any such data is perhaps even more easily led astray by their subjective experience of something, which very often does not correspond to objective reality; this is why subjective experiences are usually not regarded as being sufficient to meet the threshold for reliable knowledge or data about the world. However, the solution to both cases of error lies in education and critical thinking; and by far the best tool we have for rooting out such errors is the scientific method.
Having social and political systems informed and guided by data and individualism are important because they respect and enforce an individual’s rights and freedoms; such systems will have no need to coerce individuals into acting or behaving in particular ways, so long as they obey the law: murder, for example, or financial fraud, cannot be tolerated because they violate the freedom of others. Ignoring data and disrespecting the freedoms of individuals corrupts such systems by cutting them off from that which ensures their fair operation.
I’m guessing (hoping?) that most people reading this would surely not endorse treating someone differently based on their sex or skin-colour (and that cuts all ways). And, hopefully, the above scenario also shows how the truth about ability is not at all incompatible with treating people fairly as individuals and not judging them as members representative of a group. As I’ve tried to show, treating someone on the basis of their group identity will usually mean getting all sorts of things wrong most of the time. Treating each individual as an individual and offering him or her the opportunity to pursue whatever path is of interest to them is best, because the data backs doing just that. However, according to this study—and others like it—evidence shows that individuals having the freedom to choose which path to follow actually underscores biology-based sex differences: ‘evidence suggests gender differences in most aspects of personality—Big Five traits, Dark Triad traits, self-esteem, subjective well-being, depression and values—are conspicuously larger in cultures with more egalitarian gender roles, gender socialization and sociopolitical gender equity.’ The authors of the study go on to state: ‘Social role theory appears inadequate for explaining some of the observed cultural variations in men’s and women’s personalities.’ Nature, it would seem, is not so easily erased; indeed, why would we want to do such a thing?
Now, someone who believes in equality of outcome—rather than the equality of opportunity I’ve been discussing so far—is more than likely to dispute any data that suggests there are important differences between men and women on the grounds that almost all such differences come down to social or cultural factors. This is the second major factor in producing the ideological academic filter: blank-slatism. Blank-slatism, although it often goes undeclared, is a very common doctrine in the humanities and social sciences. Since it holds that society and culture are the primary drivers of difference and inequality, blank-slatism tends to separate people on the basis of the group to which they superficially belong and judges them on that basis. This judgement most often takes the form of a hierarchy of ‘privileges,’ which is then used to explain (and explain away) the differences identified by scientific and statistical data. Effectively, blank-slatism takes each individual to be representative of the group to which they superficially belong; so, instead of seeing overlaps in data, blank-statism reduces the individual to the group, and what results is a world with no overlaps, just rigid ‘identities.’ (This statistically illiterate process is precisely the error in thinking that drives identity politics.) Further, certain group members’ subjective experiences—usually those deemed the ‘least privileged’—are taken to be infallible, an assumption shored up by phenomenology; this formulation of subjective experience differs from the fallible individuality discussed above insofar as it can never be ‘wrong’ or ‘mistaken’ or ‘biased,’ and is inseparable from group identity. Typically (and unfortunately), any disagreement quickly gets labelled ‘racist’ or ‘sexist,’ leaving little or no ground for actual discussion to take place on.
Blank-statism must try to account for discrepancies and difference in outcome. However, as Jonathan Haidt has pointed out, the problem is that blank-slatism tends to mistake correlation for causation, another basic error in thinking that can easily be fixed by the humanities and strands of the social sciences becoming more scientifically literate. Correlation of x and y does not mean that x causes y: it may be that x causes y, but it may also be that y causes x, or z causes both x and y. Blank-slatism seizes too soon on the correlation, seeing it solely in terms of causality–e.g., ‘differences in outcome must be caused by systemic racism or sexism, etc.’–and simply stops there. In other words, having found its preferred cause, blank-slatism stops, satisfied, its thesis confirmed; meanwhile, real research and investigation continues to dig for the actual causes of the outcomes. And it is only after further research and investigation has been conducted that scientific data emerges which suggests that cross-cultural and enduring biological factors may actually be affecting outcomes.
In blank-slatism, all differences, since they are taken to be the result of social and cultural factors, are theoretically erasable. However, by adopting this position, things like data or having personal freedom of choice, since they stand in the way of the erasure, become ‘problematic’; as such, they require constant ‘interrogation,’ ‘correction’ and ‘intervention.’ So, if the reason for oppression is deemed to be x, then social and cultural factors must be actively policed for symptoms of wrongthink that is propagating x; and individuals expressing wrongthink must be relentlessly pursued and shamed. This shows how integral ‘symptomatic reading’ is to reinforcing the ideological academic filter.
It should not take much to see how putting blank-slatism into practice is necessarily authoritarian, not libertarian. In fact, this is one of the major reasons why I no longer think that the left/right axis is useful for understanding the politics of the academic ideological bubble: it is perhaps better understood using the authoritarian/libertarian axis.
EDIT: this is well worth a read.