A call for more (!) discrimination

The word “discrimination” and its variations (e.g. “to discriminate”) is hopelessly abused and misunderstood in today’s world. Indeed, while properly referring to something (potentially) highly beneficial and much needed, it has come to be a mere short for longer phrases like “sexual discrimination” and “racial discrimination”.* Worse, even these uses are often highly misleading.** Worse yet, the word has increasingly even lost the connection with these special cases, degenerating into a mere ad hominem credibility killer or a blanket term for any unpopular behavior related (or perceived as related) to e.g. race.***

*Note that it is “sexual” and “racial”—not “sexist” and “racist”. The latter two involve ascribing an intention and mentality to someone else, beyond (in almost all cases) what can possibly be known—and is sometimes manifestly false. Further, their focus on the intent rather than the criteria would often make them unsuitable even in the rare cases where the use could otherwise be justified.

**E.g because a discrimination on a contextually rational and reasonable criterion (e.g. GPA for college admissions) indirectly results in differences in group outcome, which are then incorrectly ascribed to e.g. “racial discrimination”. The latter, however, requires that race was the direct criterion for discrimination.

***Including e.g. having non-PC opinions about some group or expressing that opinion, neither of which can in any meaningful sense be considered discrimination—even in cases where the opinion or expression is worthy of disapproval. This including even the (already fundamentally flawed) concept of micro-aggressions.

What then is discrimination? Roughly speaking: The ability to recognize the differences between whatever individuals/objects/phenomena/… are being considered, to recognize the expected effects of decisions involving them, and to act accordingly. Indeed, if I were to restrict the meaning further, it is the “act” part that I would remove…* (Also see a below excursion on the Wiktionary definitions.)

*E.g. in that I would not necessarily consider someone discriminating who flipped a coin and then hired exclusively men or exclusively women based on the outcome—apart from the greater group impact, this is not much different from the entirely undiscriminating hiring by a coin flip per candidate. I might possibly even exclude e.g. the feminist stereotype of a White Old Man who deliberately hires only men because of the perceived inferiority of women: This is, at best, poor discrimination on one level and a proof of a lack of discrimination on another. C.f. below. (While at the same time being a feminist’s prime example of “discrimination” in the distorted sense.)

For instance, deciding to hire or not to hire someone as a physician based on education and whether a license to practice medicine is present, is discrimination. So is requiring a lawyer to have passed a bar exam in order to perform certain tasks. So is requiring a fire fighter to pass certain physical tests. So is using easier tests for women* than for men. So is using health-related criteria to choose between food stuffs. So is buying one horse over another based on quality of teeth or one car over another based on less rust damage. Etc. Even being able to tell the difference between different types of discrimination based on justification and effects could be seen as discrimination!

*This is, specifically, sexual discrimination, which shows that even such special cases can have the blessing of the PC crowd. It also provides an example of why it is wrong to equate “sexual” and “sexist”, because, no matter how misguided this discrimination is, it is unlikely to be rooted in more than a greater wish for equality of outcome. To boot, it is an example of poor discrimination through focus on the wrong criteria or having the wrong priorities. (Is equality of outcome when hiring really more important than the lives of fire victims?!?)

Why do we need it?

Discrimination is very closely related to the ability to make good decisions (arguably, any decision short of flipping a coin)—and the better someone is at discriminating, the better the outcomes tend to be. Note that this is by no means restricted to such obvious cases as hiring decisions based on education. It also involves e.g. seeing small-but-critical differences in cases where an argument, analogy, or whatnot does or does not apply; or being able to tell what criteria are actually relevant to understanding the matter/making the decision at hand.*

*Consider e.g. parts of the discussion in the text that prompted this one; for instance, where to draw the line between speech and action, or the difference between the IOC’s sponsor bans and bans on kneeling football players. Or consider why my statements there about employer’s rights do not, or only partially, extend to colleges: Without a lack of understanding, someone might see the situations as analogous, based e.g. on “it’s their building” or “it’s their organization”. Using other factors, the situation changes radically, e.g. in that the employer pays the employee while the college is paid by the student; that co-workers who do not get along can threaten company profits, while this is only rarely the case with students who do not get along; and that a larger part of the “college experience” overlaps with the students personal life than is, m.m., the case for an employee—especially within the U.S. campus system. (For instance, what characteristic of a college would give it greater rights to restrict free speech in a dorm than a regular landlord in an apartment building? A lecture hall, possibly—a dorm, no.)

Indeed, very many of today’s societal problems and poor political decisions go back, at least partially, to a failure to discriminate resp. to discriminate based on appropriate criteria.

Consider e.g. the common tendency to consider everything relating to “nuclear” or “radioactive” to be automatically evil (or the greater evil): Nuclear power is “evil”, yet fossil energies do far more damage to the world. The nuclear bombings of Japan were “evil”, yet their conventional counter-part killed more people. Radioactive sterilization of food is “evil”, yet science considers it safe—much unlike food poisoning… What if discrimination was not done by name or underlying technology, but rather based on the effects, risks, opportunities?

Consider the (ignorant or deliberate) failure to discriminate between e.g. anti-Islamists and anti-Muslims or immigration critics and xenophobes, treating them the same and severely hindering a civilized public debate.

Consider the failure to discriminate between school children by ability and the enforcing of a “one size fits all” system that barely even fits the average*, and leaves the weakest and strongest as misfits—and which tries to force everyone to at a minimum go through high school (or its local equivalent). (Germany still does a reasonable job, but chances are that this will not last; Sweden was an absolute horror already when I went to school; and the U.S. is a lot worse than it should and could be.)

*Or worse, is so centered on the weakest that school turns into a problem even for the average… Indeed, some claim that e.g. the U.S. “No Child Left Behind Act” has done more harm than good for this very reason.

Consider the failure to discriminate between politicians based on their expected long-term effect on society, rather than the short-term effect on one-self.

Consider the failure to discriminate between mere effort and actual result, especially with regard to political decisions. (Especially in the light of the many politicians who do not merely appear to fail at own discrimination, but actually try to fool the voters through showing that “something is being done”—even should that something be both ineffective and inefficient.)

Consider the failure to discriminate between those who can think for themselves (and rationally, critically, whatnot) and those who can not when it comes to e.g. regulations, the right to vote, self-determination, …

Consider the failure to discriminate between use and abuse, e.g. of alcohol or various performance enhancing drugs. (Or between performance enhancing drugs that are more and less dangerous…)

Consider the undue discrimination between sex crimes (or sexcrimes…) and regular crimes, especially regarding restrictions on due process or reversal of reasonable expectations. (Whether sex is involved is not a valid criterion, seeing that e.g. due process is undermined as soon as any crime is exempt from it.)

Consider the undue discrimination between Israelis and Palestinians by many Westerners, where the one is held to a “Western” standard of behavior and the other is not. (Nationality is not relevant to this aspect of the conflict.)

…

A particularly interesting example is the classification of people not yet 18 as “children”*, which effectively puts e.g. those aged 3, 10, and 17 on the same level—an often absurd lack of discrimination, considering the enormous differences (be they physical, mental, in terms of experience or world-view, …) between people of these respective ages. Nowhere is this absurdity larger than in that the “child” turns into an “adult” merely through the arrival of a certain date, while being virtually identically the same as the day before—and this accompanied with blanket rights and obligations, with no other test of suitability. Note how this applies equally to someone well-adjusted, intelligent, and precocious as it does to someone who is intellectually over-challenged even by high school and who prefers to lead a life of petty crimes and drug abuse. (To boot, this rapid change of status is highly likely to make the “children” less prepared for adulthood, worsening the situation further.)

*The size of the problem can vary from country to country, however. In e.g. the U.S. there is a fair chance that a word like “minor” will be used, especially in more formal contexts, which at least reduces the mental misassociations; in Sweden, “barn” (“child”) dominates in virtually all contexts, including (at least newer) laws.

However, there are many other problems relating to the grouping of “children” with children, especially concerning undifferentiated societal and political debates around behavior from and towards these “children”. This in particular in the area of sex, where it is not just common to use terms like “pedophile”* and “child-porn” for the entire age-range, but where I have actually repeatedly seen the claim that those sexually attracted to someone even just shy of 18 would be perverts**—despite the age limit being largely arbitrary***, despite that many are at or close to their life-time peak in attractiveness at that age, despite that most of that age are fully sexually mature, and despite that people have married and had children at considerably lower ages for large stretches of human history.

*This word strictly speaking refers to someone interested in pre-pubescent children, making it an abuse of language not covered by the (disputable) justification that can be given to “child-porn” through the wide definition of “child”. Even if the use was semantically sound, however, the extremely different implications would remain, when children and “children” at various ages are considered.

**Presumably, because the classification of someone younger as a “child” has become so ingrained with some weak thinkers that they actually see 18 as a magic limit transcending mere laws, mere biological development, mere maturity (or lack there of), and leaving those aged 17 with more in common with those aged 8 than those aged 18.

***Indeed, the “age of consent” is strictly speaking separate from the “age of maturity”, with e.g. Sweden (15) and Germany (14 or 16, depending on circumstances) having a considerably lower age of consent while keeping the age of maturity at 18.

Not all discrimination, depending on exact meaning implied, is good, but this is usually due to a lack of discrimination. Consider e.g. making a hiring decision between a Jewish high-school drop-out and a Black Ph.D. holder: With only that information present, the hiring decision can be based on either the educational aspect, the race/ethnicity aspect, or a random choice.* If we go by the educational or race aspect, there is discrimination towards the candidates. However, if the race aspect is used, then this is a sign that there has been too little or incorrect discrimination towards the hiring criteria—otherwise the unsuitability of the race aspect as a criterion would have been recognized. This, in turn, is the reason why racial discrimination is almost always wrong: It discriminates by an unsound criterion. We can also clearly see why “discrimination” must not be reduced to the meanings implied by “racial [and whatnot] discrimination”—indeed, someone truly discriminating (adjective) would not have been discriminating (verb) based on race in the above example.

*Or a combination thereof, which I will ignore: Including the combinations has no further illustrative value.

Excursion on proxy criteria:
Making decisions virtually always involves some degree of proxy criteria, because it is impossible to judge e.g. how well an applicant for a job fairs on the true criteria. For instance, the true criterion might amount to “Who gives us the best value for our money?”. This, however, is impossible to know in advance, and the prospective employer resorts to proxy criteria like prior experience, references, education, … that are likely to give a decent, if far from perfect, idea of what to expect. (Indeed, even these criteria are arguably proxies-for-proxies like intelligence, industriousness, conscientiousness, …—and, obviously, the ability to discriminate!)

Unfortunately, sometimes proxies are used that are less likely to give valuable information (e.g. impression from an interview) and/or are “a proxy too far” (e.g. race). To look at the latter, a potential U.S. employer might (correctly) observe that Jews currently tend to have higher grades than Blacks and tend to advance higher in the educational system, and conclude that the Jew is the better choice. However, seeing that this is a group characteristic, it would be much better to look at the actual individual data, removing a spurious proxy: Which of the two candidates does have the better grades and the more advanced education—not who might be expected to do so based on population statistics.

As an aside, one of my main beefs with increasing the number of college graduates (even at the cost of lowering academic standards to let the unsuitable graduate) is that the main role of a diploma was to serve as a proxy for e.g. intelligence and diligence, and that this proxy function is increasingly destroyed. Similarly, the greater infantilization of college students removes the proxy effect for ability to work and think for oneself.

Excursion on discrimination and double standards:
Interestingly, discrimination otherwise rejected, usually relating to the passage of time, is sometimes arbitrarily considered perfectly acceptable and normal. A good example is the age of maturity and variations of “age of X” (cf. above)—a certain age is used as an extremely poor and arbitrary proxy for a set of personal characteristics.

In other cases, such discrimination might have a sufficient contextual justification that it is tolerated or even considered positive. For instance, even a well qualified locker-room attendant of the wrong sex might not be acceptable to the visitors of a public bath, and the bath might then use sex as a hiring criterion. Not allowing men to compete in e.g. the WTA or WNBA can be necessary to give women a reasonable chance at sports success (and excluding women from the ATP or the NBA would then be fair from a symmetry point of view). Etc.

Then there is affirmative action…

Excursion on how to discriminate better:
A few general tips on how to discriminate better: Question whether a criterion is actually relevant, in it self, or is just as proxy, proxy-for-a-proxy, proxy-for-a-proxy-for-a-proxy, …; and try to find a more immediate criterion. Question the effectiveness of criteria (even immediate ones). Do not just look at what two things have in common (e.g. building ownership, cf. above) but what makes them different (e.g. being paid or paying). Try to understand the why and the details of something and question whether your current assumptions on the issue are actually correct—why is X done this way*, why is Y a criterion, why is Z treated differently, … Try to look at issues with reason and impartiality, not emotion or personal sympathy/antipathy; especially, when the issues are personal, involve loved ones or archenemies, concern “pet peeves”, or otherwise are likely to cause a biased reaction.

*The results can be surprising. There is a famous anecdote about a member of the younger generation who wanted to find out why the family recipe for a pot-roast (?) called for cutting off part of it in a certain manner. Many questions later, someone a few generations older, and the origin of the tradition, revealed the truth: She had always done so in order to … make the pot-roast fit into her too small pan. Everyone else did so in the erroneous belief that there was some more significant purpose behind it—even when their pans were larger.

Excursion on when not to discriminate (at all):
There might be instances where potential discrimination, even when based on superficially reasonable grounds, is better not done.

For instance, topics like free speech, especially in a U.S. campus setting, especially with an eye on PC/Leftist/whatnot censorship, feature heavily in my current thoughts and readings. Here we can see an interesting application of discrimination: Some PC/Leftist/whatnot groups selectively (try to) disallow free speech when opinions contrary to theirs are concerned. Now, if someone is convinced that he is right, is that not a reasonable type of discrimination (from his point of view)?

If the goal is to push one’s own opinion through at all cost, then, yes, it is.

Is that enough justification? Only to those who are not just dead certain and lacking in respect for others, but who also are very short-sighted:

Firstly, as I often note, there is always a chance that even those convinced beyond the shadow of a doubt are wrong. (Indeed, those dead certain often turn out to be dead wrong, while those who turn out to be right often were open to doubts.) What if someone silences the opposition, forces public policy to follow a particular scheme without debate, indoctrinates future generations in a one-sided manner, …—and then turns out to be wrong? What if the wrongness is only discovered with a great delay, or not at all, due to the free-speech restrictions? Better then to allow other opinions to be uttered.

Secondly, if the power situation changes, those once censoring can suddenly find themselves censored—especially, when they have themselves established censorship as the state of normality. Better then to have a societal standard that those in power do not censor those out of power.

Thirdly, there is a dangerous overlap between the Niemöller issue and the fellow-traveler fallacy: What if the fellow travelers who jointly condemn their common enemies today, condemn each other tomorrow? (Similarly, it is not at all uncommon for a previously respected member of e.g. the feminist community to be immediately cast out upon saying something “heretic”.) Better then to speak up in defense of the censored now, before it is too late.

Fourthly, exposure to other opinions, dialectic, eclecticism, synthesis, … can all be beneficial for the individual—and almost a necessity when we look at e.g. society as a whole, science, philosophy, … Better then to not forego these benefits.

Fifthly, and possibly most importantly, censorship is not just an infringement of rights against the censored speaker—it is also an infringement of rights against the listeners. If we were (I do not!) to consider the act against the speaker justified (e.g. because he is “evil”, “racist”, “sexist”, or just plainly “wrong”); by what reasoning can this be extended to the listeners? Short of “it’s for their own good” (or, just possibly, “it’s for the greater good”), I can see nothing. We would then rob others of their right to form their own opinions, to expose themselves to new ideas, whatnot, in the name of “their own good”—truly abhorrent. Better then to allow everyone the right to choose freely, both in terms of whom to listen to and what to do with what is heard.

Excursion on failure to discriminate in terminology:
As with the child vs. “child” issue above, there are many problems with (lack of) discrimination that can arise through use of inappropriate words or inconsistent use of words. A very good example is the deliberate abuse of the word “rape” to imply a very loosely and widely defined group of acts, in order to ensure that “statistics” show a great prevalence, combined with a (stated or implied) more stringent use when these “statistics” are presented as an argument for e.g. policy change. Since there is too little discrimination between rape and “rape”, these statistics are grossly misleading. Other examples include not discriminating between the words* “racial” and “racist”, “[anabolic] steroid” and “PED”, “convicted” and “guilty”, …

*Or the concepts: I am uncertain to what degree the common abuse of “racist” for “racial” is based on ignorance of language or genuine confusion about the corresponding concepts. (Or intellectually dishonest rhetoric by those who do know the difference…) Similar remarks can apply elsewhere.

(In a bigger picture, similar problems include e.g. euphemistic self-labeling, as with e.g. “pro-life” and “pro-choice”; derogatory enemy-labeling, e.g. “moonbat” and “wingnut”; and emotionally manipulative labels on others, e.g. the absurd rhetorical misnomer “dreamer” for some illegal aliens. Such cases are usually at most indirectly related to discrimination, however.)

Excursion on Wikipedia and Wiktionary:
Wikipedia, often corrupted by PC editors [1], predictably focuses solely on the misleading special-case meanings in the allegedly main Wikipedia article on discrimination, leaving appropriate use only to alleged special cases… A particular perversity is a separate article on Discrimination in bar exam, which largely ignores the deliberate discriminatory attempt to filter out those unsuited for the bar and focuses on alleged discrimination of Blacks and other ethnicities. Not only does this article obviously fall into the trap of seeing a difference in outcome (on the exam) as proof of differences in opportunity; it also fails to consider that Whites are typically filtered more strongly before* they encounter the bar exam, e.g. through admittance criteria to college often being tougher.**

*Implying that the exam results of e.g. Blacks and Whites are not comparable. As an illustration: Take two parallel school-classes and the task to find all students taller than 6′. The one teacher just sends all the students directly to the official measurement, the other grabs a ruler and only sends those appearing to be taller than 5′ 10”. Of course, a greater proportion of the already filtered students will exceed the 6′ filtering… However, this is proof neither that the members of their class would be taller (in general), nor that the test would favor their class over the other.

**Incidentally, a type of racial discrimination gone wrong: By weakening criteria like SAT success in favor of race, the standard of the student body is lowered without necessarily helping those it intends to help. (According to some, e.g. [2] and [3], with very different perspectives and with a long time between them.) To boot, this type of discrimination appears to hit another minority group, the East-Asians, quite hard. (They do better on the objective criteria than Whites; hence, they, not Whites, are the greater victims.)

Worse, one of its main sources (and the one source that I checked) is an opinion piece from a magazine (i.e. a source acceptable for a blog, but not for an encyclopedia), which is cited in a misleading manner:* Skimming through the opinion piece, the main theses appear to be (a) that the bar exam protects the “insiders” from competition by “outsiders” by ensuring a high entry barrier**, (b) that this strikes the poor*** unduly, and (c) that the bar exam should be abolished.

*Poor use of sources is another thing I criticized in [1].

**This is Economics 101. The only debatable point is whether the advantages offset the disadvantages for society as a whole.

***Indeed, references to minorities appear to merely consider them a special case of the “poor”, quite unlike the Wikipedia article. To boot, from context and time, I suspect that the “minorities” might have been e.g. the Irish rather than the Blacks or the Hispanics…

Wiktionary does a far better job. To quote and briefly discuss the given meanings:

1. Discernment, the act of discriminating, discerning, distinguishing, noting or perceiving differences between things, with intent to understand rightly and make correct decisions.

   Rightfully placed at the beginning: This is the core meaning, the reason why discrimination is a good thing and something to strive for, and what we should strive to preserve when we use the word.

2. The act of recognizing the ‘good’ and ‘bad’ in situations and choosing good.

   Mostly a subset of the above meaning, with reservations for the exact meaning of “good”. (But I note that even a moral “good” could be seen as included above.)

3. The setting apart of a person or group of people in a negative way, as in being discriminated against.

   Here we have something that could be interpreted in the abused sense; however, it too could be seen as a subset of the first item, with some reservation for the formulation “negative way”. Note that e.g. failing to hire someone without a license to practice medicine for a job as a practicing physician would be a good example of the meaning (and would be well within the first item).

4. (sometimes discrimination against) Distinct treatment of an individual or group to their disadvantage; treatment or consideration based on class or category rather than individual merit; partiality; prejudice; bigotry.

   sexual or racial discrimination

   Only here do we have the abused meaning—and here we see the central flaw: The example provided (“sexual or racial discrimination”) only carries the given meaning (in as far as exceeding the previous item) when combined with a qualifier; dropping such qualifiers leads to the abuse. “Sexual discrimination”, “racial discrimination”, etc., carry such meanings—just “discrimination” does not. This makes it vital never to drop these qualifiers.

   Similarly, not all ships are space ships or steam ships, the existence of the terms “space ship” and “steam ship” notwithstanding; not all forest are rain forests; sea lions are not lions at all and sea monkeys are not even vertebrates; …

   Note that some of the listed meanings only apply when viewed in the overall context of the entire sentence. Bigotry, e.g., can be a cause of discrimination by an irrelevant criterion; however, “sexual discrimination”, etc., is not it self bigotry. Prejudice* can contain sexual discrimination but is in turn a much wider concept.

   *“Prejudice” is also often misunderstood in a potentially harmful manner: A prejudice is not defined by being incorrect—but by being made in advance and without knowing all the relevant facts. For example, it is prejudice to hear that someone plays in the NBA and assume, without further investigation, that he is tall—more often than not (in this case), it is also true.

5. The quality of being discriminating, acute discernment, specifically in a learning situation; as to show great discrimination in the choice of means.

   Here we return to a broadly correct use, compatible with the first item, but in a different grammatical role (for want of a better formulation).

   I admit to having some doubts as to whether the implied grammatical role makes sense. Can the quality of being discriminating be referred to as “discrimination”? (As opposed to e.g. “showing discrimination”.) Vice versa for the second half.

6. That which discriminates; mark of distinction, a characteristic.

   The same, but without reservations for grammatical role.