The following deal is from Kit Woolsey's "Taking Control" in the August, 2000, Bridge World.

South reached four spades after East's one heart opening, and West led the heart queen. Kit suggests that East signal with the ten as suit preference. The ten cannot show attitude, Kit reasons, because East could always overtake to play hearts or spades himself.

I can't agree. If your agreement is that you play attitude at trick one with certain well-defined exceptions (as I do and as I know Kit does), you are begging for an accident if you make an ad hoc decision that, this time, your card should be suit preference. Yes, partner should draw the inference that you could overtake and defend passively if you knew that was right. But that inference merely affects the urgency of your signal, not its meaning.

Suppose you held the same hand with the diamond eight instead of the ten. Kit argues that, with that hand, you should overtake to prevent partner from shifting in a layout such as,

But how are you supposed to defeat the contract if partner has,

6 Q J 7 3 J 10 2 Q J 10 9 7

or even,

6 4 Q J 7 4 3 4 2 Q J 10 9

If you think East should not ask for a diamond shift with (B), surely it is better not to have to burn your bridges to send that message. Simply play the heart ten (attitude!). Partner now has the option of overruling you. Your signal states that you think a diamond shift is wrong, but partner knows you aren't sure, since you could prevent the shift by overtaking.

One might argue that, while my method allows you to express varying degrees of tolerance for a diamond shift, it offers no way to ask for a club shift, which Kit's method (assuming partner is on the same wavelength) does. True, but this problem exists in most trick-one situations. My philosophy is that the times you want partner to switch to dummy's potential source of tricks are rare, so you shouldn't cater to them. That doesn't mean you can never get those hands right. Sometimes you have an alarm clock signal available. And sometimes, when you suggest a passive defense by discouraging the obvious shift, partner can get it right by old-fashioned logic. He knows you have to way specifically to ask for dummy's suit, so, he should be alert to the situations where shifting to dummy's suit is as good a passive defense as any and might be productive.

Afterthoughts

Kit's article hit a nerve. I don't know how anyone can play a method where a high card sometimes asks partner to lead a diamond and sometimes specifically asks him not to, depending on subtle contextual clues. I would rather not play suit-preference signals at all than live in fear of partner's springing one on me in a situation when I would normally expect a different kind of signal. I have a number of other articles on this theme that are not yet posted. Stay tuned.