How do we know if something is good, without something lousy to compare it with?
That, colloquially expressed, is the notion of variability – it’s difficult to separate the good from the bad, when everything we see looks about the same.
Analysts take this notion to heart, and occasionally beyond conventional bounds. When one analyst I knew decided to marry, he made variability a centerpiece of his spousal-selection process. He identified his desired criteria; he dated women meeting those criteria; he dated women not meeting those criteria (for validation purposes); he used the resulting data to craft an ordered list; he proposed to women starting at the top; he was eventually accepted; he married. And not least of all, he reported his findings in an email, extolling the virtues of analytically-based life partner identification. Another acquaintance, when provided with a ridiculously small whiteboard in her cubicle, decided that if she only had room on her board for one message, it should be Give me variability, or give me death!
In formal data analysis we need variability to get anywhere, and it has a role in productive general discourse also. I was thinking about this again after reading this article, on a decrease in alumni donations to colleges. Some interesting anecdotes were offered: a donor who stopped giving when students at his alma mater shut down a conference on free speech; another donor who stopped giving when he felt prejudicially characterized by students as old, white, bigoted, and probably sexist.
These incidents suggest a lack of variability in campus discourse, along with distorted thinking, starting with invalid reasoning from aggregates – really, little more than prejudice. I agree it is disconcerting to find this level of discourse in nominally-educated people, whom we might expect to lead rational discussion on the challenges of the future. On the other hand, these are only examples – I don’t subscribe to the idea that campus discourse has completely degraded to an exchange of slogans and insults, or that it has ever been uniformly stellar.
Variability certainly has a role in reasoned discourse. It’s difficult to identify an ideal outcome when we haven’t considered the major aspects of a problem. The legal trial system has recognized this for centuries – adversaries present their positions under definite rules, support it with evidence, and reasonable people without an immediate stake in the matter reach a decision. Most of the time, it works.
While the trial system is based on what analysts would call variability, it also recognizes that meaningful discourse, especially when dealing complex or ambiguous problems, should be structured. The court and trial judge serve in no small measure to assure that discourse is on point, and that the basis for comparing positions is valid and understood by a jury.
We don’t usually have trials as a basis for discussion of diverse positions. But the trial format does suggest that variability as a basis for rational discourse requires two things: that we acknowledge there are positions to compare, and that we have an agreed-upon metric for comparison. Many verbal wresting matches never have a chance to be good discourse, because these components are missing. For example, power-generation questions often focus on one fuel source: solar is good, carbon fuels are bad, nuclear energy is death incarnate. Unfortunately, those isolated absolutes don’t permit a policy formulation – we don’t have enough solar energy to support the current economy; all reasonable evidence says that our carbon-based economy is creating serious environmental effects; nuclear energy has no carbon footprint but creates long-term storage hazard and security issues. Realistically, we’re going to need a mix of energy sources in the short term, and a government policy that promotes long-term viability. While we haggle over abolute and isolated “right” and “wrong” positions, the environmental clock presumably continues to tick. Without identifying what we’re to compare and how, we’re not even at the starting line for good discourse, and any discussion will be pre-rational. It would be better to stop and take a cue from the legal system, and decide what we’re comparing, and how to make the comparison. I’ll grant, those are often bigger challenges than the relatively small job of weighing variability once we know how to do that.
These days, we have debates galore, and it does seem that many are pre-rational, inflamed by people who have learned that reinforcing prejudice sells, and by the instant availability of opinions to reinforce almost any notion. To a real extent, the ringleaders of these verbal mob-actions don’t want rational debate – the purpose isn’t concensus, it’s to assure that their followers have something to be against. Unfortunately, failing to engage (or working to obviate) meaningful debate offers the perfect excuse for people to do nothing – an approach of nullification that seems to be increasingly favored by government leadership. Inaction presents a genuine problem. When officials can invoke an anti-policy of not spending money, not taking action, not addressing problems, or simply opposing anyone with a positive policy idea, then our government is no longer functional. I grant that government can be hyperactive, but good government identifies issues that might benefit from collective action, makes approriate decisions, and implements the corresponding policies. Doing nothing when presented with issues like climate change and failing infrastructure is not policy. Policy is action, even though many times there may be no single “right” policy – do we deploy more nuclear energy as a short-term solution to our global carbon footprint? Part of variability-based rational discussion, for all its limitations, is to undertand there is no perfect answer – often just the best answer given what we know. Politicians who believe that the non-choice of inaction is acceptable, or a policy, may need to look over the other shoulder – it seems that people are starting to notice.