In the absence of QCD penguin contributions a measurement of the time-depen
dent asymmetry in the decay B-0(t)--> pi(+) pi(-) gives directly the weak a
ngle alpha. Several bounds have been proposed in the literature on the magn
itude of the penguin effects on this determination, the prototype of which
is the isospin bound of Grossman and Quinn. We point out that large strong
final state interactions could cause these bounds to overestimate the Teal
penguin effect. A new flavor SU(3) bound is proposed, requiring only the ch
arge-averaged branching ratios for B-0 --> pi(+)pi(-) and B-s-->K+K-, which
exactly takes into account all relevant amplitudes and electroweak penguin
effects. This bound on the penguin-induced error on the determination of t
he weak phase cu holds even without knowledge of direct CP asymmetry in the
pi(+)pi(-) channel. [S0556-2821(99)04715-3].