Separation of Church and State Homepage

Grammatical arguments against the narrow reading of the Establishment Clause.

Several accomodationist scholars have made grammatical arguments suggesting that the establishment clause should be read narrowly, i.e., as banning only government actions that aid a single religion above others. Of particular importance is the argument made by Michael Malbin, a conservative legal scholar who has offered a popular non-preferentialist interpretation of the First Amendment. Malbin's work has become something of a standard jumping-off point for accomodationists, so it's important to look at just what Malbin argues, and what his arguments prove.

Malbin's most important work on church/state separation in his Religion and Politics: The Intentions of the Authors of the First Amendment. Leaving aside his treatment of the history of the religion clauses, and his discussion of Article III of the Northwest Ordinance (which we discuss in detail elsewhere in this web site), Malbin offers the following grammatical observations on the language of the religion clauses of the First Amendment:

Laycock's response:

In a 1986 article, legal scholar Douglas Laycock offers four criticisms of Malbin's grammatical argument ("Non-preferential Aid to Religion: A False Claim About Original Intent," William and Mary Law Review, Vol. 27 (1986), pp. 875-923). Each, we think, is devastating to Malbin's position. Since several of these criticisms presuppose knowledge of material covered elsewhere in this web site, we reproduce Laycock's criticisms below, and add our own commentary and links when we think them necessary.

Here are Laycock's arguments, in order:

Here Laycock refers to the Senate's rejection of, not one, but three versions of the religion clauses that would have codified the non-preferentialist position. In place of these versions the Senate adopted one of the most expansive versions of the clauses adopted by either house (we review the history of the religion clauses of the First Amendment here). The Senate's actions here speak volumes about intent, but Malbin's argument ignores these actions and instead reaches its conclusions on Congress's inconclusive and obscure preference for "an" over "the" in the establishment clause. Malbin's decision to base his conclusions on the least conclusive available evidence seems inexplicable. As Laycock puts it elsewhere: "The 'no preference' position requires a premise that the Framers were extraordinarily bad drafters--that they believed one thing but adopted language that said something substantially different, and that they did so after repeatedly attending to the choice of language....When the record reflects a textual choice as clear as this one, only extraordinarily clear contrary evidence should persuade us not to follow the text (page 882-883)."

On page 8 of Religion and Politics Malbin notes that to ban the establishment of "a religion" would clearly indicate a non-preferentialist intent, whereas banning "religion" in general would not. Unfortunately for Malbin, the framers chose to ban "religion" in general. Accordingly, Malbin makes his non-preferentialist case by leaning on the "an" in front of "establishment." But this makes no sense: "an" modifies "establishment," not "religion." Malbin's argument, in other words, assumes that the framers believed in non-preferentialism, but passed up an easy and grammatical way of indicating that belief in favor of language that is both obscure and ungrammatical. Given the care with which Congress drafted the Bill of Rights, this is both historically and logically improbable.

We know of at least 11 versions of the religion clauses considered by houses of Congress, or by a Congressional committee. None of these versions contain the phrase "the establishment of religion." There is no evidence, in other words, that anyone was concerned to distinguish between "an" and "the." If there was any choice at all, it was between versions that clearly codified the non-preferential position and versions that talked about "religion" generally. Given this choice, Congress adopted the general version. Indeed, the version eventually approved banned even laws respecting an establishment of religion.

No commentary is necessary here, except that we think that Laycock underestimates the strength of his argument. To us it seems that Laycock's reading is vastly more intuitive than Malbin's. Nothing was more obvious about state establishments of religion at the time of the Constitution than their variedness; some states had establishments while other didn't, different religions were established, and those religions were established in very different ways. If, as Malbin argues, the intention of the First Amendment was to safeguard state establishments from Congressional scrutiny, Congress would logically have used the indefinite article to reference those establishments.

Return to home