You'll go broke if you start making cars for people who fall asleep at night dreaming about flawless transportation appliances like the Toyota Camry.

Terrorists kidnap a Japanese and an American automotive engineer and announce they both will be executed by firing squad. However, to show they are not complete barbarians, the captors give each one a last request.

The Japanese engineer says he'd like to give one last lecture on quality improvement. Upon hearing that, the American engineer requests that he be shot first.

It's an old joke from the early 1980s. Until it was told and re-told from podiums at engineering conferences all over the world, it usually got a good chuckle — even from Japanese engineers.

Everybody knew that Detroit's Big Three automakers were getting sick of hearing about how great Japanese quality was.

Judging by many of the letters we get, they still are. Despite making huge improvements from the bad old days of the '70s and '80s, General Motors Corp., Ford Motor Co. and the Chrysler Group continue to trail Japanese automakers in most independent quality surveys such as those done by J.D. Power & Associates.

And now U.S. quality is being assailed again, thanks to the soaring number of recalls during the past several years.

Besides complaining that the media are biased toward foreign products, readers — particularly engineers — often offer up examples of where Detroit's products are superior in specific categories such as braking, crash test performance, or other areas. After I praised the quality and refinement of the Toyota Sequoia SUV in an earlier issue, I got a number of letters informing me that the Sequoia had less usable room inside than its domestic competitors, got worse fuel economy and did poorly on several U.S. government parking brake tests.

Although none of these arguments substantially lowered my opinion of the Sequoia, the letters do highlight the dilemmas journalists face in writing about quality issues and engineers face in trying to improve their products.

Unfortunately, talking about vehicle quality to different segments, disciplines and companies within the auto industry often is reminiscent of the old fable of three blind men describing an elephant. Some like to talk about quality using statistics, defect rates and catch phrases like “Six Sigma.” Others talk about fit and finish. Still others talk about design, recall rates or paint gloss.

Add to that a totally unfocused concept of what quality really means to the car-buying public and you've got a difficult situation. Furthermore, as Jerry Flint points out in his column on p.19, car buyers will ignore lots of glitches if they're in love with other aspects of a vehicle. And that's absolutely true.

You'll go broke if you start making cars for people who fall asleep at night dreaming about flawless transportation appliances like the Toyota Camry. One of the most muddled statements ever to come out of the U.S. Supreme Court was when former Justice Potter Stewart tried to define obscenity by saying “I can't tell you what it is, but I know it when I see it.”

As pathetically vague as that sounds, this is usually the same criteria auto writers use when they talk about “refinement” and the “quality” of test vehicles, and the “quality” of vaguely described componentry such as “interior materials.” It has to be terribly frustrating for engineers who are looking for highly empirical analyses, but the fact of the matter is, when you test drive dozens of different vehicles every year, you do get a feel for good and bad vehicle characteristics that all relate to the overall perception of the quality of a vehicle.

This would be a lousy evaluation system if it weren't exactly the same one that consumers use when they shop for a new vehicle.

Companies such as J.D. Power & Associates have done a wonderful job of trying to scientifically quantify vehicle quality, but much of its data still is not all that illuminating. For instance, its most closely followed report, which tracks vehicle problems during the first 90 days of ownership, doesn't distinguish between big problems and small ones.

And even recalls aren't a very good measure of vehicle quality, or potential buyer dissatisfaction. Alan Hine, an independent design consultant from Ortonville, MI, is disgusted with the quality of his wife's '98 Mercedes-Benz M-Class SUV, yet one of his beefs is that Mercedes hasn't recalled the vehicle enough.

He also owns two Audi sedans, and he likes the fact that the manufacturer occasionally calls them back to replace or upgrade parts. “Most of these recalls are minor things — they're coming up with a better mousetrap,” he says. Mercedes' refusal to recall its vehicles more frequently seems like dereliction of duty to Mr. Hine, or at least arrogance, he says. Among his myriad M-Class beefs: two transmission failures that have stranded his wife, and cheap interior trim that falls off at will. Don't try to tell him the Europeans are catching up with the Japanese on quality.

So what options do Detroit automakers have left? Some of the latest research from the Center for Automotive Research in Ann Arbor, MI, suggests that Japanese automakers are excelling in quality because they focus more on how well a part or system functions in a specific role rather than how well the individual parts meet specifications.

It sounds both simple and mind-bogglingly different from the way quality is now managed at the Big Three.

Perhaps there is at least one more speech on Japanese quality that's worth a listen.

Listen to Drew Winter and other Ward's editors Monday and Thursday on WJR 760 AM radio in Detroit.