The Husks of Dead Theories
This week the Supreme Court hears Ricci v. DeStefano, the case of the 20 firefighters in New Haven, Connecticut. The plaintiffs took promotion exams five and a half years ago. As they themselves say on their website (scroll down to CASE SUMMARY):
We are a group of lieutenants and firefighters in the New Haven Fire Department who underwent a grueling multi-phase civil service examination process to qualify for promotions to the command ranks of Captain and Lieutenant. In early 2004, the city needed to fill vacancies in both ranks. Although we were deemed the most qualified candidates, we were denied the promotions because we are White or Hispanic and not enough Black candidates succeeded.
(There is a bit of squid ink there. In fact only one of the plaintiffs is Hispanic. The average scores for the Hispanic test-takers were around the 20th percentile of non-Hispanic white test takers. That is, around eighty percent of non-Hispanic whites had better scores than the average Hispanic. The performance of black test-takers was similar to that of Hispanics.)
Once the city of New Haven realised that no black test-takers, and only one Hispanic, had reached the pass mark, they threw out the test. Nobody has been promoted. This is of course unfair to the test-takers — not to mention the citizens of New Haven! — but in the convoluted state of current discrimination law, the city had little choice. If they'd promoted those who passed the test, they'd have been sued by disgruntled black and Hispanic test-takers. Having thrown out the test results to avoid this outcome, they are now being sued by white (and one Hispanic) examinees. Discrimination law gets you coming and going.
There is nothing new here, of course. Given the history of this subject, the really surprising thing is that as late as 2003 a fire department was still giving formal examinations for promotions. New York City Police Department was fighting lawsuits over "discriminatory" test results thirty years ago. Police, fire, and other municipal departments all over the country have been similarly affected across an entire generation.
Attempted solutions have included every kind of rigging and "race-norming" of results, the dumbing-down of the tests to a point where well-nigh everyone passes (candidates then being promoted by lottery or straightforward race quotas), the hiring of expensive consultants to devise bias-free tests, or just giving up on tests altogether, as New Haven has now done.
None of it helped, though just dumbing down the tests has proved fairly effective for litigation-avoidance. (In 1991 New York City Sanitation Department gave a test on which 23,078 applicants out of 24,000 got perfect scores — try spotting a race gap there!) The careful concocting of scrupulously bias-free tests is now a profitable specialty within the management consulting field. New Haven hired the Houston firm of Jeanneret & Associates, Inc., who called in a contractor named I/O Solutions to devise firefighter tests, and the city spent over a hundred thousand dollars in fees to these firms.
It did no good, of course. It never does. New York Police Department spent ten years trying to write tests for promotion to Sergeant that would pass court approval. Concluding at last that any test that required reading would deliver results biased against minorities, they worked up a series of video-based tests. Just 1.7 percent of African American test-takers passed them.
The unhappy fact is that different ethnic groups exhibit different profiles of results on tests. Attempts to devise a test on which this does not happen have all failed, across decades of effort, criticism, and analysis.
Nobody knows why this is so; but the fact that it invariably, repeatedly, and intractably is so, makes testing hazardous, ultimately pointless, under current employment law. Yet still employees must be selected somehow from applicant pools, and there must be some clear, fair criteria for their subsequent promotion. The state of the law now is that almost anything an organization does in this area will open it to litigation.
Ricci v. DeStefano takes place in a time of general public exhaustion over racial inequalities. We'd really rather just not think about it. Fifty years ago it all seemed cut and dried. Just strike down old unjust laws, give the minority a helping hand, give the non-minority some education about civil rights and past disgraces, and in a few years things will come right.
We coasted along under those assumptions for a generation. When it became obvious that things were not coming right in the matter of equal test results, scholars and jurists got to work on the problem.
Liberals, with their usual coarse stupidity, naturally assumed it was just a matter of spending more money on schools. This theory was tested to destruction in several places, most sensationally in Kansas City from 1985-97. Under a judge's order, the school district spent two billion dollars over twelve years, pretty much rebuilding the school system — and the actual schools themselves — from the ground up. The new, lavish facilities included "an Olympic-sized swimming pool with an underwater viewing room, television and animation studios, a robotics lab, a 25-acre wildlife sanctuary, a zoo, a model United Nations with simultaneous translation capability, and field trips to Mexico and Senegal." The experiment was a complete failure. Drop-out rates rose and test scores fell across the entire twelve years. (Here are current test scores for the school that got the Olympic-sized swimming pool. I could not find any published results for achievement in aquatic sports.)
Conservatives, thoroughly race-whipped by the liberal media elites, preferred to go along with whatever liberals said, except that they made, and still make, mild throat-clearing noises about school vouchers. It has turned out in practice, however, that the only people keen on school vouchers are the striving poor, a small (and dwindling) demographic with no political weight, and whom nobody in the media or academic elites gives a fig about. The non-striving underclass has zero interest in education; middle-class suburbanites like their schools the way they are, thanks all the same; and teacher unions see vouchers as threats to the public-education gravy train their members ride to well-padded retirement.
As test gaps persisted and lawsuits multiplied, the scholars retreated into metaphysics. The word "culture" was wafted around a lot. It seemed to denote a sort of phlogiston or luminiferous aether, pervading and determining everything, but via mechanisms nobody could explain. We heard about self-esteem issues, "the burden of 'acting white'," "stereotype threat," and a whole raft of other sunbeams-from-cucumbers hypotheses. Stephan and Abigail Thernstrom, two distinguished scholars in the field, produced a much-praised book about test-score gaps with a conclusion in which nothing was concluded. "Choice [of where to live] should not be a class-based privilege." Where, in a free society, has it ever not been? How will you stop people moving, if they can afford to? "Families must help their children to the best of their ability." Oh. "Vouchers are a matter of basic equity." See above. "Big-city superintendents and principals operate in a bureaucratic and political straitjacket." True, no doubt; but New Haven, pop. 124,000, is not a big city. Test-score gaps are in plain sight out here in the 'burbs. John Ogbu wrote a book about it. Six years ago.
And the test-score gaps just sat there, and sat there, and sat there, grinning back at us impudently.
At last, we just stopped thinking about the whole disagreeable business. Unfortunately, by that time a great body of law had been built on the theories and pseudotheories of the preceding decades, and couldn't be wished away. Hence Ricci v. DeStefano.
You can deduce our state of exhaustion from booksellers' lists. I just spent half an hour trawling through the bibliographies and references of my own modest collection of social-science literature to come up with the following list of 50 published books, most by accredited scholars, relevant to Ricci v. DeStefano and the issues underlying the case. I offer it to the Supremes as a reading list, if they'd like to get up to speed on the necessary sociology.