Originalism and Textualism in Action: Not Constraining and Not Neutral
by Joseph Kimble
In an August 25, 2017 blog post, Michael Dorf asked, “How Determinate Is Originalism in Practice?” His conclusion from the evidence: not very.
There followed a series of exchanges between Prof. Dorf and Prof. Lawrence Solum about sample size, alternative reasons for originalists’ ideologically conservative results (such as the need to sometimes compromise with the nonoriginalist justices), the reluctance of progressives to make originalist arguments, and other counters to Prof. Dorf’s conclusion. In the end, I thought his conclusion stood up: while you might expect originalism to lead to moderately conservative results, the decisions of Justices Scalia and Thomas are anything but. Even without data, it’s “blindingly obvious” (as Prof. Dorf puts it) that their decisions are overwhelmingly conservative. Originalism does not constrain.
And neither does its companion, textualism. Here the empirical evidence is compelling, if not incontrovertible.
A central principle of both originalism and textualism is the “fixed-meaning canon,” as Justice Scalia and Bryan Garner call it in their book Reading Law (p. 78): “Words must be given the meaning they had when the text was adopted.” Textualism, the authors assert (p. xxviii), “will curb — even reverse — the tendency of judges to imbue authoritative texts with their own policy preferences.” And the authors inveigh against (p. 16) “the slander that [textualism] is a device calculated to produce socially or politically conservative results.” Rather, “[a] textualist reading will sometimes produce ‘conservative’ outcomes, sometimes ‘liberal’ ones.”
However true that may be in theory, it is not true in practice — especially in the cases that matter.
In a 2015 article (pp. 30–35), I summarized six empirical studies: four show a strong ideological bent in Justice Scalia’s opinions; one other concludes from an analysis of over 600 Supreme Court cases that the textual canons of construction “are regularly used in an instrumental if not ideologically conscious manner”; and the last one concludes that the constraining effect of originalism “is overstated at best and illusory at worst.” Could all these studies have been wrong or flawed?
In that same article (p. 35, note 96), I cited 11 other scholarly sources that cast doubt on the neutrality and consistency of Justice Scalia’s textualism.
If you want one more piece of empirical evidence to add to the pile, I offer my recent article on overrulings by the Michigan Supreme Court between 2000 and 2015. The court’s majority for almost all that time consisted of justices who were nominated by the Republican party or appointed by a Republican governor. All of them were exponents of textualism. And since they were the majority, they did not need to compromise.
I did a detailed study of 96 (!) overrulings by the court during those years. I coded the civil cases according to whether they made it harder to sue and recover or get relief, or made it easier to sue. I coded the criminal cases according to whether the overruling favored the prosecution by making it easier to convict, harder to get a charge dismissed, harder to appeal, harder to get a resentencing, and so on — or whether the overrulings favored the defendants in some way. Of the 96 overrulings, 15 did not fit these categories, leaving 81.
The results would not have surprised Michigan lawyers, but they ought to shock everyone else. In 46 of 48 civil cases, Republican majorities voted in a way that makes it harder for plaintiffs to successfully sue. (Think tort reform.) In 32 of 33 criminal cases, Republican majorities favored the prosecution. In all: civil plaintiffs and criminal defendants lost in 78 of 81 overrulings, for a conservative ideology rating of 96.3.
You can argue about rough measures. You might quibble with a few of the codings. But you can hardly deny the general import of those numbers: they cannot be rationalized by any jurisprudence that’s evenhanded. And if you were to argue that the court was simply correcting an ideological tilt of previous courts (majority decisions going back decades and even centuries), that in itself would be an argument grounded in policy, not text.
Incidentally, I examined four of the overruled cases to show how close they were and how the disputed language was not nearly as clear and plain as the textualist majority claimed. It staggers credibility to think that previous majorities could have misread text so many, many times.
The upshot is this (p. 376): “In practice, textualism has devolved into a vehicle for ideological judging — disguised as deference to the legislature. The numbers in Michigan, though, blow its cover.”
In an August 25, 2017 blog post, Michael Dorf asked, “How Determinate Is Originalism in Practice?” His conclusion from the evidence: not very.
There followed a series of exchanges between Prof. Dorf and Prof. Lawrence Solum about sample size, alternative reasons for originalists’ ideologically conservative results (such as the need to sometimes compromise with the nonoriginalist justices), the reluctance of progressives to make originalist arguments, and other counters to Prof. Dorf’s conclusion. In the end, I thought his conclusion stood up: while you might expect originalism to lead to moderately conservative results, the decisions of Justices Scalia and Thomas are anything but. Even without data, it’s “blindingly obvious” (as Prof. Dorf puts it) that their decisions are overwhelmingly conservative. Originalism does not constrain.
And neither does its companion, textualism. Here the empirical evidence is compelling, if not incontrovertible.
A central principle of both originalism and textualism is the “fixed-meaning canon,” as Justice Scalia and Bryan Garner call it in their book Reading Law (p. 78): “Words must be given the meaning they had when the text was adopted.” Textualism, the authors assert (p. xxviii), “will curb — even reverse — the tendency of judges to imbue authoritative texts with their own policy preferences.” And the authors inveigh against (p. 16) “the slander that [textualism] is a device calculated to produce socially or politically conservative results.” Rather, “[a] textualist reading will sometimes produce ‘conservative’ outcomes, sometimes ‘liberal’ ones.”
However true that may be in theory, it is not true in practice — especially in the cases that matter.
In a 2015 article (pp. 30–35), I summarized six empirical studies: four show a strong ideological bent in Justice Scalia’s opinions; one other concludes from an analysis of over 600 Supreme Court cases that the textual canons of construction “are regularly used in an instrumental if not ideologically conscious manner”; and the last one concludes that the constraining effect of originalism “is overstated at best and illusory at worst.” Could all these studies have been wrong or flawed?
In that same article (p. 35, note 96), I cited 11 other scholarly sources that cast doubt on the neutrality and consistency of Justice Scalia’s textualism.
If you want one more piece of empirical evidence to add to the pile, I offer my recent article on overrulings by the Michigan Supreme Court between 2000 and 2015. The court’s majority for almost all that time consisted of justices who were nominated by the Republican party or appointed by a Republican governor. All of them were exponents of textualism. And since they were the majority, they did not need to compromise.
I did a detailed study of 96 (!) overrulings by the court during those years. I coded the civil cases according to whether they made it harder to sue and recover or get relief, or made it easier to sue. I coded the criminal cases according to whether the overruling favored the prosecution by making it easier to convict, harder to get a charge dismissed, harder to appeal, harder to get a resentencing, and so on — or whether the overrulings favored the defendants in some way. Of the 96 overrulings, 15 did not fit these categories, leaving 81.
The results would not have surprised Michigan lawyers, but they ought to shock everyone else. In 46 of 48 civil cases, Republican majorities voted in a way that makes it harder for plaintiffs to successfully sue. (Think tort reform.) In 32 of 33 criminal cases, Republican majorities favored the prosecution. In all: civil plaintiffs and criminal defendants lost in 78 of 81 overrulings, for a conservative ideology rating of 96.3.
You can argue about rough measures. You might quibble with a few of the codings. But you can hardly deny the general import of those numbers: they cannot be rationalized by any jurisprudence that’s evenhanded. And if you were to argue that the court was simply correcting an ideological tilt of previous courts (majority decisions going back decades and even centuries), that in itself would be an argument grounded in policy, not text.
Incidentally, I examined four of the overruled cases to show how close they were and how the disputed language was not nearly as clear and plain as the textualist majority claimed. It staggers credibility to think that previous majorities could have misread text so many, many times.
The upshot is this (p. 376): “In practice, textualism has devolved into a vehicle for ideological judging — disguised as deference to the legislature. The numbers in Michigan, though, blow its cover.”