A Reply to Five Friends

• Date: 01 September 2016
• Author: Frederick Schauer
• DOI: http://doi.org/10.1111/raju.12132
• Published date: 01 September 2016

A Reply to Five Friends

FREDERICK SCHAUER*

Abstract. For an academic, there is no greater reward than having one’s scholarship

taken seriously. Thefive distinguished scholars who have contributed to this sympo-

sium on The Force of Law (Schauer 2015) have done just that, with varying degrees of

agreement and disagreement, praise and criticism. But even critical commentary,

and perhaps especially critical commentary, is evidence of serious engagement.

More importantly, thecommentaries contained here have advancedour understand-

ing of law in valuable ways. I respond to each in this reply, but with full acknowl-

edgment that my responses cannot do justice to the full breadth of their

contributions and challenges. My hope is not that I will persuade readers that I am

correct and my critics mistaken, but rather that the reader who absorbs both the

challenges and my response will come away with a greater understanding of the

issues that The Force of Law seeks to place on the agenda of contemporary

jurisprudence.

1. Leslie Green, Na€

ıve Empiricism, and the External Point of View

Leslie Green takes issue with several of my claims, and his points are often well-

taken. Some of his sound observations are exegetical matters about H. L. A. Hart’s

Concept of Law (Hart 2012), and others go to broader jurisprudential questions. Lit-

tle would be served by listing the instances in which his comments have exposed

gaps in my own argument or presentation,

1

but in one important respect a genuine

disagreement remains, and it is one that goes to the heart of jurisprudential method

and substance. More particularly, Green (2016) accuses me of a “na€

ıve empiricism,”

* The symposium of whichthis reply is part was held at Bocconi University, Milan, on October

2, 2014, and appearsin the previous edition of thisJournal. I am grateful to Damiano Canaleand

Giovanni Tuzet, both of Bocconi University, and to Nicoletta Ladavac, of Centre Themis,

Geneva, forconceiving, planning, sponsoring,and conducting such a stimulatingevent.

1

I do think it importantto note, however, that Green and I have a genuineand important exe-

getical disagreement about the purpose of Hart’s (2012, 40) reference to the “puzzled man.”

Green argues that Hartuses the phrase as part of a response to Kelsen and not to Holmes, and

that Hart does so in order to makethe point that law is more than just directives to or empow-

erments of officials. But given that the “puzzled man”phrase appears only two sentences after

a reference to the “badman,” and given that the “bad man” phrase comes from Holmes (1897,

458) and not from Kelsen, it is hard to avoid the conclusion that Hart’s target is at least as

much Holmes and his prediction account of law as it is Kelsen and his directives-to-officials

account.

V

C2016 The Author. Ratio Juris V

C2016 John Wiley & Sons Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main

Street, Malden 02148, USA.

Ratio Juris. Vol. 29 No. 3 September 2016 (348–363)

but it is not entirely clear what form of empiricism he is describing, what makes it

na€

ıve, and whether I am guilty of it.

One interpretation of Green’s charge is that it is na€

ıve, or simply mistaken, to

engage in empirical examination without having a clear specification of just what

one is examining. Or, to put it somewhat closer to Green’s language, it is a mistake

to try to count things without knowing just what one is counting.

2

And here the

issue is whether it is possible to count instances of behavior, or anything else, if

what one is seeking to count has core and fringe examples, or is a cluster connected

only by a family resemblance, or in some other way cannot be fully specified in

advance of the counting. Green appears to believe it is mistaken to try to even

examine the presence or frequency of the items we wish to investigate without

more of a specification of the nature of some entity than cluster, prototype, or fam-

ily resemblance accounts of our concepts would allow. Others, however, insist that

this kind of precise specification is unnecessary at the outset, even if the result of a

conceptual or empirical inquiry might then enable further specification.

3

But

although the question of just how much specification by way of essential properties

is necessary to engage in causally-directed empirical investigation presents an

interesting and important methodological debate, it may not be a debate that is as

relevant to my use of empirical evidence as Green supposes. Although I plead

guilty to a belief that essential-feature accounts of law, coercion, and much else are

misguided, most of the empirical evidence marshaled in The Force of Law is in sup-

port of a claim about which I do not deny the existence of a pretty clear under-

standing of just what it is that we are trying to measure. More particularly, the

empirical findings to which I refer and on which I rely are directed to a precise and

well-specified issue on which Green and I agree: To what extent have officials and

ordinary citizens internalized law qua law as a reason for their decisions?

4

At the

heart of The Force of Law is the claim that such internalization, absent force, coer-

cion, sanctions, or other forms of external incentives, is far rarer than is commonly

supposed. And so although I agree with Green that knowing what you are trying

2

To describe empirical research as “counting” seems slightly pejorative, although I do not

know whether the mild denigration was Green’s goal. But good empirical research, like some

of the research on which I draw, is not in the business of counting. Rather, it seeks to iden-

tify causal connections between phenomena, as with the causal connections between law and

judicial or nonjudicial decisions and judgments. It is true that examining causality in this

way requires some assessment of the frequency of both the independent and dependent vari-

ables, and if this is all that Green means then we have no disagreement on this score. But if

he is suggesting that good empirical research is about counting as an end in itself, then I

demur.

3

Roger Shiner, for example, engagingly uses the example of electric toasters to illustrate that

what we need to identify something as a toaster—or law—is far less than a full account of the

nature of toasters—or law (Shiner 1989,10–15; 1992, 5–9, 130–32). To thesame effect, see Halpin

1998, 187–205; Leiter 2007, 2011, 663–77. Indeed, we might think of the relationship between

specificationand empirical research as a sort of reflectiveequilibrium process, such thatwe pre-

liminarily specify the independent and/or dependent variables, reach a tentative research con-

clusion about causation, and then use thattentative conclusion to further specify the variables,

and then produceanother conclusion about causation, and then engagein further iterations in a

process thatmay never come to a complete end.

4

And for this purpose, as I say in the book, Gavison’s (1987, 30–1) simultaneously crude and

helpful ideaof “first stage law” is pretty muchall we need.

A Reply to Five Friends 349

Ratio Juris, Vol. 29, No. 3 V

C2016 The Author. Ratio Juris V

C2016 John Wiley & Sons Ltd.