Thursday, August 6, 2015

Normative versus positive arguments

From Are Biden-for-President Supporters All Sexist? by Jonathan Chait.

Chait argues:
One of the unfortunate habits overtaking the left is a tendency to conclude that any behavior that could plausibly be motivated by bigotry is likely motivated by bigotry. It is no doubt true that a misogynist would want Joe Biden to challenge Hillary Clinton. Therefore, Scott Lemieux concludes, people who want Biden to challenge Clinton are sexist.
He concludes:
The point is not that Biden would make a better nominee than Clinton. (As I said, I don’t even think that.) The point is that reasonable people can disagree about this. It’s not as though Biden were some obscure figure suddenly thrust forward in a desperate search for a plausible white male nominee. He’s the two-term vice-president of the United States. Sitting vice-presidents run for the top job all the time. Lemieux's argumentative method is to insist that, since he doesn't consider Biden more electable than Clinton, nobody could consider Biden more electable than Clinton, leaving only sexism as a plausible account for their beliefs.

This particular form of illogic has gotten endemic on the left. A racist would oppose Barack Obama, but that doesn't make all opposition to Obama racist. Likewise, a sexist would hate Hillary Clinton, but maybe we shouldn't spend the next 15 months — and possibly the subsequent four or eight years — defining all opposition to her as sexist.
I think Chait is correct but I would couch it in a larger framework than simple left:right.

I have no way of measuring this but I have the sense that there has been a change in the balance in public debates between the two philosophical categories of normative arguments and positive arguments. From Wikipedia:
In philosophy, normative statements make claims about how things should or ought to be, how to value them, which things are good or bad, and which actions are right or wrong. Normative claims are usually contrasted with positive (i.e. descriptive, explanatory, or constative) claims when describing types of theories, beliefs, or propositions. Positive statements are (purportedly-) factual statements that attempt to describe reality.

For example, "children should eat vegetables", and "those who would sacrifice liberty for security deserve neither" are normative claims. On the other hand, "vegetables contain a relatively high proportion of vitamins", and "a common consequence of sacrificing liberty for security is a loss of both" are positive claims. Whether or not a statement is normative is logically independent of whether it is verified, verifiable, or popularly held.
Normative versus positive is a separate issue from rhetorical versus logical/empirical.

I sense that over recent years (since the emergence of the internet?) there has been a greater comfort in relying on normative arguments over positive arguments. I think the proportions might have changed. Chait might be, and indeed, I suspect is, right that this shift is manifested to a greater degree on the left than on the right (see some of Jonathan Haidt's work in this area). My point is that normative arguments in a heterogenous society are far less productive than positive arguments and therefore the rise of the proportion of arguments that are simply normative is counterproductive to effective decision-making.

Normative arguments are essentially a form of a religious argument, an appeal to non-human authority. Gun control, abortion, campus rape, global warming, education reform, etc., these are all characterized by an increasing appeal to normative beliefs over positive arguments. As an example of the barrier that this poses, consider the pattern of national debate after any of the mass public killings in recent years such as Chattanooga, Charleston, Newport, etc. After each incident, there is routinely a call from gun advocates to increase gun control. But in virtually every instance, the proposed solutions, had they been in place a priori, would not have changed the outcome.

This is pure normative argument. Gun control because we should have gun control. The positive argument is far harder to make, and the implications far more uncomfortable, which is why I think it is more rarely made: what are the actions we can take within our Constitutional parameters and existing circumstances which are likely to decrease the incident rate and lethality of mass public killings? This isn't about the normative belief about gun control, this is something much more useful, what can we do that will make a difference? By relying on normative arguments, we forestall actual progress.

There are a couple of other items sparked by Chait's observation. The left does have a strong inclination to raise charges of racism or gender discrimination (normative arguments) in reply to any opposing positive argument. That inclination is both ineffective (a positive argument can only be addressed by an alternative positive argument) and counterproductive (it is the equivalent of "Shut up, I don't agree!"). Calls for conversations are disingenuous if normative argumentation is to be the only basis of discussion. I think these issues are why there is the common characterization that Liberals think Conservatives are evil (normative) whereas Conservatives think Liberals are stupid (positive arguments).

The other thought arising from Chait's observation is that this inclination to make baseless normative charges is part-and-parcel with a belief about the implications of disparate impact. Different groups (with different goals, priorities, abilities, behaviors, norms, backgrounds, contexts, etc.) are disparately affected by new policies and different groups have disparate outcomes. The Left has come, in the past few decades, to look at disparate impact as ipso facto evidence of discrimination when it has never been anything other than a possible indication. Or, as Chait puts it, there is a belief that
any behavior that could plausibly be motivated by bigotry is likely motivated by bigotry.
But that is simply wrong. The causes of disparate impact are innumerable and while there are occasional instances where bigotry and discrimination can be the cause of the disparate impact, in most instances, the variance is traceable to non-contentious causes.

Take a simple example where two groups, Group A and Group B are exactly alike in all material ways and have exactly the same incomes and exactly the same propensity to save. The only way in which they differ is that Group A, on average, only have a single child each generation, while Group B have four. Now look forward three generations. They still have the same incomes and saving rates and still look exactly the same in every way except one, Wealth. With Group A, each generation's accumulated wealth is passed to the next in a concentrating fashion. Group B generations receive a declining proportion of generational wealth owing to the number of children born to each family. Group A, with great concentrations of accumulated wealth appear to be the beneficiaries of some systemic favor when in fact, it is the outcome only of differences in fertility rates.

Now, change one single attribute of identity which has no impact on anything else. Make them two different races, or ethnic groups or religions or from different regions or classes, etc. Still the same in all their other behaviors (except fertility). Someone, myopically focused on identity such as race, and convinced that disparate impact can only arise from discrimination, will look at Group A and Group B and conclude that the difference between them can only have arisen from discrimination rather than being the consequence of differences in fertility practices.

That is the grave disservice of focusing on identity and of accepting the premise that disparate impact can only be the result of discrimination. By so censoring alternative explanations, you miss what is actually happening and thereby fail to make the changes that are actually going to make a positive change.

In today's environment, a person of the left would look at Group A and decide that a redistributive wealth tax was called for to equalize opportunities for all groups. Group A would lose much of their accumulated wealth and Group B would be the beneficiaries of that coercive largesse. And in three generations, the Wealth Gap would be back to where it was because familial fertility practices, the actual root cause of the variance, were not changed. Such is the consequence of focusing on normative arguments rather than positive arguments.

My take away is that this is a cultural issue applicable across the board. We need to reacquaint ourselves with the ability to make positive arguments. Normative arguments work much better in an environment of cultural conformity where everyone shares a common belief set. Absent that, positive arguments are the only way to make progress. By focusing on positive arguments, 1) you reduce the number of instances of baseless claims of racism or gender discrimination, 2) conversations are not shut down as often and different groups are not as easily isolated from one another, and, best of all, 3) actual progress can be made tackling real problems.

No comments:

Post a Comment