Thats Just What It Is To Be Legally Valid by accinent


									                      That's Just What It Is To Be Legally Valid:

             A Quick Refutation of Greenberg's "How Facts Make Law"

1.     Introduction

In a recent paper1 Mark Greenberg offers a novel argument against legal positivism.

Legal practices, argues Greenberg, determine the legal facts (like which rules are

legally valid). But this determination is not merely metaphysical determination.

Rather, it is rational determination, the kind of determination that shows why the

metaphysical determination is intelligible, the kind of determination that is non-

arbitrary, that explains why the legal facts should depend on the legal practice in some

specific way rather than some other one. But legal facts cannot rationally determine

themselves, and neither can purely descriptive facts. Something else is needed for

rational determination. And the best candidates for this role are normative, or value

facts. So value facts play an indispensable role in determining legal facts. So legal

positivism is false. So argues Greenberg.

2.     Rational Determination and Identity

Clearly, then, the idea of rational determination plays a crucial role in Greenberg's

argument. And indeed, Greenberg himself thinks of the introduction of this notion as

one of the major contributions he is making to the jurisprudential literature, which has

for many years now been about the determination relation, but has said nothing

explicit about what this determination is (158). Given these facts, one would expect to

see much more from Greenberg by way of a characterization of this relation. His

semi-official characterization of this relation is somewhat cryptic: The A facts

rationally determine the B facts, says Greenberg, "just in case the A facts

metaphysically determine the B facts and the obtaining of the A facts makes

intelligible or rationally explains the B facts' obtaining." (163), but he has almost

nothing to say on what "making intelligible" or "rationally explaining" is. And though

at other places he says quite a few things about the rational determination relation

(many of which I list in a footnote2), he does not offer a detailed, reasonably explicit

and precise account of what this relation comes to3. I think this is not just a minor

flaw – for I think that there is no way of filling in the details here that will preserve

whatever plausibility Greenberg's argument seems to have.

        To see this, think of a positivist view according to which legal validity is

determined only by some complex social facts, say facts about what officials in

certain positions (descriptively, non-legally characterized) recognize as binding on

them. Furthermore, according to this view, being recognized by such officials is just

what it is to be legally valid. This sketched view, I take it, is well within the scope of

the intended target of Greenberg's argument. But nowhere in his original paper (and

hardly anywhere else) does Greenberg explicitly discuss the following question: Does

the that's-just-what-it-is relation qualify as an instance of rational determination?4

Given the multiplicity of metaphors Greenberg uses to give a feel for what he has in

mind when talking about rational determination5 and the vagueness of his official

characterization of this relation, the answer is not at all clear. Does a true that's-just-

what-it-is claim render the higher-level facts or their relation to the determining facts

intelligible? Does a that's-just-what-it-is claim qualify as a rational explanation?

Could such metaphysically necessary identities be opaque, in the intended sense of

"opaque"? After all, though the belief in such an identity statements call for epistemic

justification, the identity fact itself doesn't seem to be the kind of fact that calls for

explanation6. Furthermore, such identity statements are often – as in the case of the

legal positivist, at least of the kind sketched above – supposed to be a priori, so if a

priori entailment is the paradigm of rendering intelligible (165), a that's-just-what-it-is

claim should qualify as an instance of rational determination. And it's not clear what

we can say about reason-talk when it comes to such identity claims. Furthermore, it

seems like perfectly rational creatures should be able to know such identity facts

(especially if they are apriori). If so, there is nothing opaque about a that's-just-what-

it-is relation, and so perhaps we should conclude that this relation is (or at least can

be) an instance of rational determination. On the other hand, Greenberg repeatedly

contrasts (164) cases of rational determination with cases of simple reduction, where

presumably some kind of identity relation does crucial work7. So perhaps we should

conclude that a that's-just-what-it-is relation does not suffice for rational

determination after all.

       The fact that the sketched positivist view is hardly an anomaly among

positivist theories, and the naturalness of putting things in terms of a that's-just-what-

it-is relation, already serve to emphasize the need for more details: If we are to be

given an argument against positivism with such a central role for the idea of rational

determination, we had better be given an account of rational determination that can

clearly yield an answer to the simple yes-no question: Can a that's-just-what-it-is

relation qualify as an instance of rational determination8?

3.     Greenberg's Argument Refuted

But things get even worse. For we can now refute Greenberg's argument using the

following dilemma9.

        Either a that's-just-what-it-is relation suffices for rational determination (in the

sense Greenberg wants to give to this phrase), or it doesn't. If it does, then

Greenberg's argument fails against the sketched positivist view, and any other that

utilizes a that's-just-what-it-is relation in its specification of the relation between legal

and social facts. This leaves it open that the argument works against some other

versions of positivism10, but it seems to me the most influential and plausible ones are

the ones that escape Greenberg's argument (thus understood). Be that as it may, at

least as a general argument against positivism – and this is how the argument is

intended – the argument clearly fails.

        On the second horn of the dilemma, we are to assume a specification of the

rational determination relation such that a that's-just-what-it-is relation cannot suffice

for rational determination. Of course, there is nothing objectionable in so

understanding the expression "rational determination" (which is, after all, Greenberg's

invention). But thus understood, the premise that legal practices rationally determine

legal facts amounts to the claim that there is no identity between legal and social facts.

And while this premise may be true, assuming it in an argument that purports to refute

positivism is an obvious case of begging the question. Why would a positivist (or

even one who is an agnostic about the positivism debate) accept a premise that says,

in effect, that legal facts are not identical with social facts?11

        Greenberg owes us, then, many more details about his rational determination

relation. But he has no way of filling them in while preserving the force of his

argument: If he fills in the details in a way that allows for a that's-just-what-it-is

relation to be an instance of rational determination, then positivism (in its common,

most plausible versions) satisfies the rational determination requirement, and is thus

off the hook. And if Greenberg fills in the details in a way that rules out a that's-just-

what-it-is relation as an instance of rational determination, then his argument may (for

anything I thus far said) be sound, but only at the price of being clearly question-

begging12. Either way, then, as an argument against positivism the argument fails13.

Indeed, it only seems plausible upon a first reading of Greenberg's text because of the

unclarity about the nature of rational determination.14

       Now, at one point in a follow-up paper (2006b, 127) Greenberg mentions

(under the title "Hartian Positivism") a view that seems to be committed to a that's-

just-what-it-is relation between legal and social facts, and he even seems to wonder

whether on such a view legal practices rationally determine legal facts15. But all he

says in reply is that – as he argues at length in another paper (2006a) – this kind of

Hartian Positivism "is not part of our reflective understanding of law". Now, I have no

idea whether this is so. But this doesn't matter for our purpose here (namely, that of

evaluating Greenberg's argument presented in the opening paragraph). For if an

argument can be given with the conclusion that Hartian Positivism is "not part of our

reflective understanding of law", then it is this argument that refutes positivism, and

not the one we're here discussing, the one where the idea of rational determination is

supposed to do serious work16. And if the only way of showing that the rational

determination argument works as against a central kind of positivism is by assuming

that there is some other argument that refutes that kind of positivism, then the rational

determination argument again begs the question against the legal positivist, and does

no dialectical work at all.

4.     Nevertheless, Supporting the Rational Determination Premise?

On the second horn of the dilemma (a that's-just-what-it-is relation does not suffice

for rational determination), Greenberg's argument is, as I've just argued, question-

begging. But perhaps things would be different if the crucial premise – legal practices

rationally determine (in a sense that rules out as insufficient a that's-just-what-it-is

claim) legal facts – could be supported on independent, non-question-begging

grounds. So perhaps this is what Greenberg should do.

        But notice, first, that this would be an entirely different argument. The new,

improved argument will have the claim that legal practices rationally determine legal

facts as a conclusion, rather than as a premise. Indeed, with such an argument at hand,

the concept rational determination is likely to prove redundant, so that whatever

argument can support the rational determination premise can directly refute

positivism. And of course, if it is such an argument that is needed (because the

original argument is question-begging), then Greenberg can no longer afford the

nonchalance (170; 2006b, 118) of explicitly offering only an initial defense of the

rational determination premise – everything depends on the reasons that can be

offered for this premise, and so a full defense is badly needed here. And so he most

certainly can't argue that the rational determination premise – understood as ruling out

a that's-just-what-it-is relation – "should be uncontroversial" (169).

        Lastly, the preliminary support Greenberg does offer for the rational

determination claim is fairly quick and unconvincing, at least when we are clear about

what this claim is to mean (namely, that a that's-just-what-it-is relation is insufficient

for rational determination). Greenberg uses examples, such as "facts about democracy

and fairness explain the relevance of the publicly promulgated decisions of elected

legislators to the content of the law." (2006b, 118); he repeatedly pumps intuitions

about it not being arbitrary which particular social facts determine which particular

legal ones; he relies on a disanalogy between the legal case and the case of the funny

or the beautiful, where opaqueness seems unproblematic (160); he sketches (170-1) an

epistemological argument according to which if the determination relation were not

intelligible, we would not be able to know the legal facts; and in general he seems to

argue that our practices of legal reasoning embody a commitment to rational

determination (170-172).

       But none of this succeeds in supporting rational determination (when this

relation is understood, remember, as ruling out a that's-just-what-it-is relation as

insufficient for rational determination). The examples ("facts about democracy and

fairness explain the relevance of the publicly promulgated decisions of elected

legislators to the content of the law.") don't help, because what is intuitive about them

can be easily explained in other ways: For instance, facts about democracy and

fairness help explain why it is a good idea for the publicly promulgated decisions of

elected legislators to have a certain legal effect. Similarly, within a legal system in

which ideas about democracy and fairness already have some legal status (some rules

expressing them are already legally valid), then the kind of explanation Greenberg

mentions can be understood as an intra-legal explanation. Needless to say, both these

ways of accommodating the intuitive force of the examples are perfectly available to

the positivist. It's hard to see that anything in this example (or in related ones) can

support Greenberg's conclusion – namely, that the facts about democracy and fairness

explain why, as a matter of legal fact, the publicly promulgated decisions of elected

legislators are relevant in a specific way to the content of the law. For similar reasons,

talk of arbitrariness is misleading: Given the availability of intra-legal explanations of

the kind just mentioned, and of normative explanations of why it's a good idea for the

law to be determined in a certain way by the social facts, arbitrariness can be avoided

(and so the intuitions about non-arbitrariness can be accommodated) without anything

like rational determination.

        The disanalogy between the legal case and the aesthetic one is indeed

interesting – in the aesthetic case we typically have less to say about why some

(presumably physical) determining facts determine the relevant specific aesthetic

facts. But this disanalogy does not support the rational determination premise: In the

legal case we often need (and are able) to say more, because of the kinds of

consideration already mentioned (intra-legal considerations, and normative ones about

what rules should be legally valid). It's just that at some point we are stuck with a

that's-just-what-it-is claim. Nothing in the disanalogy with the aesthetic case (and

related cases) counts against this picture.

        As for Greenberg's epistemological argument: Given the suggested that's-just-

what-it-is claim, this epistemological argument becomes really a challenge to the

knowability of (at least a priori) identities. It is an interesting challenge, but not one

we need to deal with here, because, first, it's a much more general challenge, and

second, and consequently, though it's not clear how it can be met, it's very clear that it

can be met, for we do know many identities of the relevant kind.

        As for our legal reasoning practices and the commitments that are embedded

in them: There is undoubtedly a commitment to give and ask for reasons. But we've

already seen how such a commitment can be accommodated consistently with

rejecting rational determination. In order to support rational determination (as we are

understanding it in this section), Greenberg would have to show that our legal

reasoning practices involve a commitment to never settle with a that's-just-what-it-is-

to-be-legally-valid claim. He hasn't done so, and I find it very hard to believe that this

can be done.

5.      Conclusion

To conclude, then: Greenberg's argument is crucially unclear and incomplete because

Greenberg fails to fill in the details of what he is after by the stipulated term "rational

determination". Furthermore, there is no way of filling in the details here without

rendering the original argument either powerless against most common and plausible

kinds of positivism, or question-begging. If non-question-begging support could be

offered for the rational-determination premise, perhaps another argument could be

constructed, one that is not unrelated to Greenberg's project. But this would be a

different argument than Greenberg's official argument, he does not offer convincing

support for this premise, and it is hard to see how such support could be offered.

        Legal positivism may yet be false, of course. But Greenberg's argument does

not, I conclude, make progress in showing that this is so.


Mark Greenberg, (2004), "How Facts Make Law", Legal Theory 10, 157-198.

Mark Greenberg, (2006a), "Hartian Positivism and Normative Facts: How Facts Make

        Law II", in Exploring Law's Empire (Scott Hershovitz ed.); available here:

Mark Greenberg, (2006b)"On Practices and the Law", Legal Theory 12, 113-136.

Ram Neta, "On the Normative Significance of Brute Facts", Legal Theory 10 (2004),


Mark Schroeder, "Cudworth and Normative Explanations", Journal of Ethics and

        Social Philosophy 1(3) (2005), available on line at

Kevin Toh, "An Argument Against the Social Fact Thesis (And Some Additional

           Preliminary Steps Towards a New Conception of Legal Positivism)", Law and

           Philosophy 27 (2008), 445-504.

    Greenberg (2004). All page references in the text are to this paper, unless otherwise noted. For more

of the same project, see Greenberg (2006a, 2006b).
    The rational determination relation is "primarily a metaphysical, or constitutive, one, and only

secondarily an epistemic one" (158), but unlike bare metaphysical determination, it is "necessarily

reason-based" (160), and this notion of a reason "may well be best understood as an epistemic notion"

(160), so we may "have an epistemic notion playing a role in a metaphysical relation" (160); in cases of

rational determination it has to be in principle intelligible why certain determining facts determine

certain determined facts (160); in cases of rational determination, as opposed to bare metaphysical

determination, a perfectly rational being must be able to see why it is that particular A facts make

particular B facts obtain (164); in explicitly metaphorical terms, "the relation between the law practices

and the content of the law must be transparent" (164); a priori entailment from A facts to B facts is "a

clear example of rational determination", for if the relation is a priori, it is "afortiori intelligible" (165);

"It is not possible that the truth of a legal proposition could simply be opaque, in the sense that there

would be no possibility of seeing its truth to be an intelligible consequence of the law practices" (170);

Rational intelligibility concerns not the content of the higher-level facts … but the relation between the

determining facts and the higher-level facts." (2006b, 132).
    Neta (2004, 202) registers a similar complaint.
    Though he discusses the possible generalizability of his argument to other areas (2006), Greenberg

doesn't discuss the question whether his argument can be applied to other rule-governed practices, like

games. When asked why a chess-rules fact is a chess-rules fact, at some point we are going to have to

answer something like: Well, that's just what it is to be playing chess. Our positivist is giving the

analogous answer when it comes to law.
    See footnote 2, above.
    With regard to bare metaphysical necessity, Greenberg says: "To dramatize the point, even a perfectly

rational being may not be able to see why it is that particular A facts make particular B facts obtain."

(164). But this way of putting things is highly misleading, for it suggests that in those cases there may

be a good why-question to which the perfectly rational being has no answer. But, of course, if an

identity claim is involved, then it seems like there is no good why question to begin with.
    Could it be that the distinction Greenberg is really after is just that between a priori and a posteriori

identity statements?
    In a response to Greenberg (2004), Neta (2004) uses the example of what makes breaking a promise

wrong, and attempts an answer in completely descriptive terms. In his reply, Greenberg (2006b, 125)

rightly notes that Neta illegitimately takes for granted that facts about desires explain or give rise to

normative reasons. But Greenberg does not notice that if having a desire is just what it is to have a

normative reason, then Neta's move becomes respectable again. (I have no idea whether this is a

commitment Neta wants to undertake.)
    My argument's form is similar to Neta's (2004) main line of reply to Greenberg, as he too presents his

reply as a dilemma of sorts that starts with different possible understandings of "rational

determination". But the details of the argument in the text are importantly different from Neta's, and as

a result, nothing in Greenberg's reply to Neta (2006, 209 and on) applies, as far as I can see, to the

argument in the text here.
     Although, of course, there may be other flaws in Greenberg's arguments, flaws I do not discuss here.

And depending on the details of the relevant positivist view and of the rational determination relation,

it's quite possible that other kinds of positivist views can escape Greenberg's argument in a way

precisely analogous to that in the text.
     In a very recent paper (2008), Kevin Toh suggests to distinguish – in the context of the debate over

positivism – between internal and external legal statements. Toh accepts a social identity thesis with

regard to the latter, but offers a presumptive argument against it with regard to the former. I think that

this move is subject to serious objections which I hope to discuss elsewhere. Let me just note here that

Greenberg nowhere seems to restrict his argument to just internal legal statements.
     At one point (2006a, 23) Greenberg accuses his interlocutor who attempts to rely on an identity

statement as begging the question against Greenberg. But Greenberg seems to misunderstand the

dialectical situation. He (Greenberg) attempts to offer a novel argument against legal positivism. We

then find out that if the novel argument is to have any force at all against the most common and

plausible kind of positivism, its major premise must be interpreted in a way that will not allow any

positivist (or agnostic) to accept it. This is sufficient to show that Greenberg's novel argument fails

(because it is question-begging). This, of course, does not suffice to establish positivism (relying on

positivism to do that would, of course, constitute question-begging), but this is not what we're trying to

do. Rather, we're trying to evaluate the force of one (Greenberg's) argument against positivism. And if

the argument relies on a premise that says, in effect, that positivism is false, then it is without force.
     Nor will it help Greenberg to try and escape the horns of my dilemma by saying that some cases of a

that's-just-what-it-is relation satisfy the rational determination premise, and others don't. For we can

then ask whether the case of the positivist view sketched in the text is a case of the former or the latter,

and proceed with the dilemma as before.
     I suspect that behind the scenes in Greenberg's discussion there is another, deeper and more

interesting confusion, a confusion about how normative explanations work. For an elaborate and

intricate discussion of this general confusion (not in the jurisprudential context), and for the claim that

some normative explanations can be constitutive (in my terms here, they can rely on a that's-just-what-

it-is claim), see Schroeder (2005).
     At times, Greenberg talks about the invocation of such an identity statement as an explanation of

rational determination in terms of "some truth specific to the legal case" (2006a, 15). This way of

putting things is misleading, though. True, the specific identity claim invoked is of course specific to

the legal case (it is, after all, a claim about what legal validity consists in). But the pattern of explaining

such things in terms of an identity claim applies generally.
     And indeed, in the section where Greenberg attempts to show that Hartian Positivism is "not part of

our reflective understanding of law" (2006a, Section V) the idea of rational determination does no work

at all, and Greenberg relies instead on other familiar lines of thought that are sometimes taken

(unconvincingly, I think) to pose problems for legal positivism.


To top