314 ANALYZING DATA
As noted, there are many more fruitful and illuminating approaches to critically assessing the
value assumptions and claims of advocates involved in public policy disputes, but these should
suffice for illustrating what is meant by a “value-critical” analysis. In each case, an effort is made
to systematically interrogate the assumptions and value claims made by protagonists in the policy
conflict under study. The point I want to stress here is that this form of critical analysis is interpre-
tive and involves making judgments, but this does not mean that the criteria for judgment and
interpretation are without standards of validity. So long as the analyst makes explicit her or his
reasoning and the standards being used for judgment and interpretation, readers are able to re-
spond critically as well, facilitating the kind of dialogic process that is essential for clarifying—
and sometimes resolving—public policy disputes.Step Five: Drawing Conclusions; Making RecommendationsThe final step in this form of value-critical policy analysis involves drawing conclusions from
one’s analysis and possibly making policy recommendations as well. Drawing conclusions from
a value-critical analysis involves making as clear as possible just what is at stake, for whom, in a
public policy dispute. That is, having critically assessed the strengths and weaknesses of each
major protagonist’s core arguments and value assumptions, the analyst needs to provide an over-
view of the larger meanings and significance of the analysis itself. What, finally, does this dispute
come down to? How can we summarize, finally, what is at stake, for whom, in this public policy
conflict? And this is not just a matter of restating what one has written before; rather, what is
needed here is a distillation, a clarifying reformulation of the meaning and significance of the
analysis itself.
In my value-critical analysis of the U.S. language policy debate, I concluded with the argu-
ment that what is finally at stake in this debate is not principally about language at all, but that
rather the stakes are concerned with “identity politics,” with what kind of country we are, and,
more specifically, with the relationships between multiple levels of human identity: individual,
ethno-linguistic group, and nation. Is the United States fundamentally a “monistic” society when
it comes to language and culture, or is the country a “pluralistic” society in relation to these forms
of human identity? This question ultimately directs us to the fundamental stakes involved in the
U.S. language policy debate, and assimilationists and pluralists are in fundamental disagreement
in relation to this question. Resolving this disagreement would go a long way toward resolving
the language policy debate; it cannot be resolved at the level of language use alone. And though
my value-critical analysis faults both sides in the debate, it was on the basis of this analysis that I
came to the judgment that pluralists provided a better argument overall than did assimilationists.
In addition to providing a concluding analysis of the public policy dispute, the value-critical
analyst might also want to offer recommendations for how the policy argument could best be
resolved. That is, in light of the findings of the value-critical analysis, it may be appropriate for
the policy analyst to suggest the shape of a preferred public policy on the issue at hand. This, of
course, will depend on the role of the analyst. In the event that a city council employed someone
(or assigned a staff member) to do a value-critical analysis of an important issue, its members
might wish to see only the critical analysis of major value assumptions and arguments made by
protagonists in a policy debate, and not care to hear the recommendations of the staff member. It
is conceivable though that other policy makers might want the analyst to offer well-reasoned
recommendations as well.
As noted above, academics such as the present author are in a social role that enables us to
make recommendations that we believe flow directly from our analyses. In the case of my analysis