Faigin cites three papers, the most recent of which is the 1990 paper by
Westphal that John presented in his post.  The abstract of one of the other
two papers, pasted below, claims a 28% relative area under the insulin
curve above baseline.

Both of these studies have small sample sizes of "normal subjects," seven
or eight.  It'd be useful to know how much variation was caused by the test
conditions and by the subjects themselves.

The effect of protein ingestion on the metabolic response to oral glucose
in normal individuals.

Am J Clin Nutr 1986 Dec;44(6):847-56   (ISSN: 0002-9165)

Krezowski PA; Nuttall FQ; Gannon MC; Bartosh NH [Find other articles with
these Authors]

Eight normal subjects were given 50 g protein, 50 g glucose, or 50 g
protein + 50 g glucose. Plasma glucose, insulin, C-peptide, glucagon, alpha-
amino nitrogen (AAN), and nonesterified fatty acid (NEFA) responses were
then determined over 4 h. Protein stimulated only a modest insulin rise and
the area above fasting baseline was only 28% of that after glucose. The sum
of the serum insulin area following protein ingestion and that following
glucose ingestion was 100.4% of the combination meal. C-peptide changes
confirmed the insulin response. The addition of glucose to the protein meal
resulted in a 60 min delay in glucagon and AAN rise compared to the protein
meal alone. Subsequently AAN and glucagon increased to levels greater than
or equal to those observed after protein ingestion alone. In summary,
protein is a much less potent secretagogue for insulin than is glucose in
normal individuals, and the effect on insulin secretion is not synergistic.
Addition of glucose to a protein meal results in a delayed rise in AAN and
glucagon concentrations in normal subjects.