The New Natural Law Theory as the Source of Bostock’s Error

by James Berquist

Bostock vs. Clayton: The Arguments of Gorsuch and Alito in Brief

If you have ever wondered what practical significance the understanding or misunderstanding of the natural law presented by the New Natural Law (NNL) theorists might have in public life, look no farther than the strange arguments presented in the majority’s opinion in Bostock vs Clayton.

Neil Gorsuch, a student of John Finnis (a founder and chief proponent of NNL theory), argues the following:

If the employer intentionally relies in part on an individual employee’s sex when deciding to discharge the employee—put differently, if changing the employee’s sex would have yielded a different choice by the employer—a statutory violation has occurred… An individual’s homosexuality or transgender status is not relevant to employment decisions. That’s because it is impossible to discriminate against a person for being homosexual or transgender without discriminating against that individual based on sex.[1]
… Consider, for example, an employer with two employees, both of whom are attracted to men. The two individuals are, to the employer’s mind, materially identical in all respects, except that one is a man and the other a woman. If the employer fires the male employee for no reason other than the fact he is attracted to men, the employer discriminates against him for traits or actions it tolerates in his female colleague. Put differently, the employer intentionally singles out an employee to fire based in part on the employee’s sex, and the affected employee’s sex is a but-for cause of his discharge.[2]

He completes his argument by means of another example:

A model employee arrives and introduces a manager to Susan, the employee’s wife. Will that employee be fired? If the policy works as the employer intends, the answer depends entirely on whether the model employee is a man or a woman. To be sure, that employer’s ultimate goal might be to discriminate on the basis of sexual orientation. But to achieve that purpose the employer must, along the way, intentionally treat an employee worse based in part on that individual’s sex.[3]

So, to sum it up: to discriminate on the basis of sexual orientation necessitates that one discriminate on the basis of sex, because if—in a thought experiment—the same inclination were present while the sex of the employee were changed, then the reason for discrimination would appear or disappear, respectively. Robert loves Susan (and openly lives in accordance with that inclination), and the employer is fine with that, but if Robert were actually Roberta, then the employer would fire her. It is all about the sex of ‘Robert(a).’

Alito is more than surprised at this line of argumentation and, in particular, the last example cited:

This example disproves the Court’s argument because it is perfectly clear that the employer’s motivation in firing the female employee had nothing to do with that employee’s sex. The employer presumably knew that this employee was a woman before she was invited to the fateful party. Yet the employer, far from holding her biological sex against her, rated her a “model employee.” At the party, the employer learned something new, her sexual orientation, and it was this new information that motivated her discharge.[4]

Yes, but Gorsuch would say that while the information was new, it was only relevant because of the sex of the employee. So, Gorsuch would say, it all comes down to the sex, and is therefore in violation of Title VII.

The key question, as both Alito and Gorsuch understand it, is what motivates the firing of the employee. Alito says it is the sexual orientation that motivates the firing, Gorsuch says that it is the sex plus the orientation that so motivates, and all that is needed (to be in violation of Title VII) is a “but-for” condition. Even while the employer focuses on the same sex orientation, the fact remains that if the sex of the employee had been different, so also would the employer’s attitude.

Alito backs up his explanation with the following argument:

The problem with this argument is that the Court loads the dice. That is so because in the mind of an employer who does not want to employ individuals who are attracted to members of the same sex, these two employees are not materially identical in every respect but sex. On the contrary, they differ in another way that the employer thinks is quite material. And until Title VII is amended to add sexual orientation as a prohibited ground, this is a view that an employer is permitted to implement.[5]

So, Alito is saying, the difference between a man attracted to women (erotically, and acting from that attraction) and a woman so oriented is not simply that one is a man and the other a woman, but rather that one is erotically inclined to members of the same sex, and the other is erotically inclined to members of the opposite sex. Gorsuch, Alito is implying, is treating the ‘erotic inclination to women’ as if it is predicated univocally of both men and women, but it is not. Thus, the motivation is not about the sex, it is about the different kinds of erotic orientation.

Alito’s charge is absolutely correct. It would be helpful, however, to explain more clearly why one cannot predicate the particular orientation univocally of men and women. Gorsuch’s mistake, which is absolutely typical of the NNL theorists in general and Finnis in particular, is that he attends to reality as if things existed in reality in the same way they exist in thought. In our thoughts, we can abstract and separate things that cannot be separated in reality. The NNL theorists and, again, Finnis in particular tend to treat of things separated in thought as if they could be separated in reality. This can be a danger in speculative understanding, to be sure, but it is deadly in questions of motivation, action, and practical reasoning in general. For we can separate good things in thought and then treat of their goodness or desirability as if they could also be parceled out, as it were, according to the separations or abstractions in our mind.

This must be further explicated. I do so by analyzing Gorsuch’s particular error, and then showing the same kind of error in the principles of the NNL theory, and Finnis’ work on fundamental human goods specifically. Finally, I show the danger to any judicial theory subject to these errors.

Gorsuch’s Particular Error Examined

Let me give an example parallel to Gorsuch’s “fateful party” example. Let us suppose that Robert invites his boss to his home for a party and introduces his boss to his ‘spouse,’ x. Is x male or female? The fate of Robert depends upon the answer. To use Gorsuch’s mode of interpretation, this means that the action of the employer—and thus the motivation of the employer—will spring from the sex of the spouse. If ‘x’ is Susan, the employer will be fine and happy. If ‘x’ turns out to be Charles, the employer will be motivated to fire Robert. Thus, but for the sex of x, Robert would be fine. It seems that the motivation for firing depends upon sex, specifically the sex of the spouse.[6]

I don’t offer this example facetiously, or to score any cheap points. I am sure Gorsuch’s response would be that the sex of Robert was just as informative of the situation. But in reframing the example slightly, I am pointing out the kind of argumentation that Gorsuch is employing. He is saying that the motivation of the action on the part of the employer comes down to the information that leads to the understanding of the situation. The employer learns of the orientation of his employee by understanding the sex of his employee plus coming to know the sex of his employee’s ‘spouse.’

This is true as stated, but the point in question is not what information leads to an understanding of the whole situation, but rather what is the reality of the situation, and in particular what is the reality of the love that is being talked about. I may say that I do not understand that a person is living a homosexual lifestyle until I see that he is erotically inclined and acting upon that inclination to individual men. I thus see that he is homosexual when I consider his sex plus his erotic inclination and action upon this inclination to other men. But at that moment, my understanding of ‘attracted to men’ is differently said of this situation than it would be in the case of a woman ‘attracted to men.’ And that is because I hold that erotic love is only naturally ordered when it is between men and women. Thus, while my understanding of the reality comes from the putting together of terms in the way Gorsuch sets out, my understanding of the terms themselves changes as they come together. I cannot predicate/say ‘attracted to men’ in the same way of men and women unless I am merely attending to the reality according to the abstraction and separation in my mind. Thus, the employer really can be motivated by the orientation as opposed to the sex, to complete Alito’s argument because it is not important whether the employee is a man or woman; what matters to the employer is the reality of erotic love, and he wants his employees to live in accordance with that reality.

Gorsuch is doing something the NNL theorists do consistently (if usually more broadly) and especially in their accounts of the principles of human agency. He is analyzing a situation abstracted from concrete reality, separating terms, and then saying that the agent is motivated according to this abstraction and separation of terms. Gorsuch thus considers a situation wherein a man is attracted to men (in the relevant ways) and separates the terms “man” and “attracted to men” as if the understanding of what it means to be ‘attracted to men’ is something that can be understood so abstracted and applied univocally in reality. But ‘attracted to men’ does not exist in a vacuum; it only exists in this or that individual, or you can say it exists in the mind of one considering it abstractly. However, the point is that it does not exist in reality in the same way that it exists in the mind. In the latter, it is predicable of either men or women, and thus it exists in the mind abstracted from proper or improper orientation. By itself, it indicates neither. That is why it can be said of both. But it exists in men (and in this man) as an improper orientation—i.e. not in accordance with nature, and in women (and in this woman) as a proper orientation—i.e. in accordance with nature. It is therefore the inclination as it exists in the subject that is pertinent. The employer is concerned not about the sex, but the proper or improper orientation of the individual employee regardless of sex.

To get technical,[7] this is why logic and metaphysics (and also natural philosophy) are not the same science—in the scholastic sense—and why logic is an art rather than a science. Logic concerns the existence things have in the mind and the production of intentions, statements, and syllogisms, while metaphysics considers being as such (and natural philosophy considers natural being specifically). For the logical intentions exist in the mind and we incline to the reality through logical intentions. And thus while we use them in every science, we must recognize when we are considering the reality in itself, or considering it according to the logical abstraction through which we incline to the actual subject. It is not bad to do the latter as we do it of necessity as the above examples indicate, but we must not confuse the abstraction’s existence with the reality’s existence, or the kind of predication that works univocally in abstraction with the analogical or equivocal predication that obtains in reality. This is the case in both speculative and practical matters, but perhaps there is a special urgencyin practical matters where what is important to the agent is what his action is about in reality.

Gorsuch’s argument is practical (in kind) inasmuch as he is talking about the way reason shapes action. Again, he argues that because the employer would act differently based upon whether or not the employee is a man or woman who lives out an erotic attraction to women or a man, therefore the reason for the employer’s action in part comes down to sex. But this is to claim that action is rooted not in being as it is in things, but rather as it is in the mind. While it is true to say that we cannot desire what we do not know, and to this extent it must be in our mind, it is foolish to say we want things according to the existence they have in the mind. When we incline to a thing, we want it, not the thought of it. When we make something (and this the NNL is better about), we want the being of the thing in reality and not merely in our mind. But the being that we want (or don’t want) is always the existence outside of the mind. As Aristotle and Thomas consistently say, goodness is in things.

So, to close off this section on Gorsuch’s error, we should emphasize that the employer who is concerned that his workplace respect the natural order, and therefore wishes to prevent any scandal in regards to the true ordination of erotic love, is not concerned with the sex of his employees even in part. He is focused upon the good of the natural law and order regardless of a person’s sex. Whether or not the employee is a man or a woman, the employer simply requires the same reality be respected.

John Finnis and the (lack of a) hierarchy of goods

In the next section, I detail the general version of Gorsuch’s error that permeates the most basic principle of the NNL theory. To do so, let’s first examine an identical mode of argumentation and confusion in John Finnis’s work. In his influential book, Natural Law and Natural Rights, John Finnis argues that there is not an intrinsic hierarchy amongst the basic or fundamental goods of human life.

To understand his position adequately, we should first explain what is meant by “basic” or “fundamental human goods” in NNL theory. Inasmuch as something can be listed as desirable for its own sake such that one need not reference anything further in order to explain its desirability, that something is called a basic or fundamental human good. Christopher Tollefsen gives a nice summary of the basic goods position of the NNL theorists:

… practical reason, that is, reason oriented towards action, grasps as self-evidently desirable a number of basic goods. These goods, which are described as constitutive aspects of genuine human flourishing, include life and health; knowledge and aesthetic experience; skilled work and play; friendship; marriage; harmony with God, and harmony among a person’s judgments, choices, feelings, and behavior. As grasped by practical reason, the basic goods give foundational reasons for action to human agents. Moreover, they are recognized as good for all human agents; it is equally intelligible to act for the sake of the life of another as for one’s own life.[8]

The actual list of goods has been something the NNL theorists have gone back and forth on, but they remain consistent in claiming that the basic goods are those that can be reasons for action all by themselves. Anything that is desirable for its own sake (such that one need look no further for a reason for action) is a basic good or the instantiation of one.

John Finnis argues, in line with the rest of the NNL theorists, that these goods have no intrinsic order amongst themselves, since they can all be ‘good for their own sakes.’ And here is how Finnis defends the position in a critical passage in Natural Law and Natural Rights:

If one focuses on the value of speculative truth, it can reasonably be regarded as more important than anything; knowledge can be regarded as the most important thing to acquire; life can be regarded as merely a precondition, of lesser or no intrinsic value; play can be regarded as frivolous; one’s concern about ‘religious’ questions can seem just an aspect of the struggle against error, superstition, and ignorance; friendship can seem worth forgoing, or be found exclusively in sharing and enhancing knowledge; and so on. But one can shift one’s focus. If one is drowning, or, again, if one is thinking about one’s child who died soon after birth, one is inclined to shift one’s focus to the value of life simply as such. The life will not be regarded as a mere precondition of anything else; rather, play and knowledge and religion will seem secondary, even rather optional extras. But one can shift one’s focus, in this way, one-by-one right round the circle of basic values that constitute the horizon of our opportunities.[9]

It all depends on what one focuses on in the moment or a given set of circumstances. Each basic good can be seen to be more important than the others, or perhaps one set over another. To present a parallel example, if one witnesses another person drowning, in that moment, the goodness of life will seem more important than the goodness of knowledge. But in a moment of persecution, one might recognize that one should be willing to stand up for the worth and goodness of truth even if it means risking or losing one’s life.

But this mode of argumentation—while presenting the truth in part—is exactly like Gorsuch’s insofar as he tries to make a conclusion about the order of goodness by an analysis according to the existence things have in the mind. His example—where one’s analysis of the situation rests upon certain concepts separated from others in thought (health and the further perfection of truth)—shows that one thinks about the situation through these concepts, and then concludes that in the given situation under consideration, one’s desire and motivation for action is in accordance with that separation. In the struggle for life that occurs when one is drowning, one aims at the goodness of life, and probably doesn’t even think about the goodness of knowledge in any explicit way. Yet, if I take a step back, is it not evident that even the drowning man desires life and health for what it is in reality and not according to the kinds of separation that occur in his mind? He may not attend to the fact that being alive and healthy makes him able to pursue the truth (or any number of other goods—indeed, all of them!) but what he wants in reality is a condition for the pursuit of these other more perfect goods, for that is what life is. If we recognize, as we should, that man is ordered to the contemplation (knowledge and love) of the supreme truths—for it is man’s intellect and rational will that distinguishes him from all other creatures in the visible world—then we see that life and health are ordered to the knowledge and love of truth. There is an order in reality, and we desire goods according to that reality.

Whatever Finnis makes of Gorsuch’s argument (and I suspect that NNL theorists could still object to Gorsuch’s argument on other grounds, such as claiming that this sort of ruling violates the human ordination to all the basic goods—especially to the good of marriage—and also the genuine political harmony required for pursuing the basic goods), it is evident that Gorsuch is falling prey to the same kind of confusion found in the Finnis example.

The broader confusion in NNL principles

To be clear, the problem here is not what I will call the ‘logical analysis’ of the situation. Our actions are rooted in such analysis inasmuch as we come to know the situation in light of these separations in thought. This is what makes the NNL approach potentially more dangerous than most errors; it is so close to the truth in a certain respect. For it is true that in a moment of decision the agent will categorize and compare his options through such separations in thought. If I am thinking about eating some ice cream, I will likely in some respect weigh the good of health over against the good of sweetness and pleasure, etc. In that moment, the desirability of health can be considered apart, as Finnis says, from the goodness of knowledge or marriage or harmony amongst my inner passions, etc. But none of this analysis actually explains the motives of action, for it does not address the kind of desire. I may desire health itself for all sorts of reasons, or at least three, finally.[10] I may desire it because it is useful for many other activities, as Finnis first intimates. Or I may desire it because it is pleasurable (for instance, when one feels ill and in pain, the desire for health may be mostly about a desire to be free from pain). Or, I may recognize that my life according to my nature—and thus also all human life—and all other life in its own measure is a share in the divine, as Aristotle puts it,[11] and thus something to be loved for its own sake. The point is that the kinds of desire are not identical with distinctions in the kinds of things. We desire many things and can desire them for diverse reasons. The basic goods are rooted in a division of things, but diverse motivations for action are finally explained according to a proper division of goods as such.

As one becomes more familiar with the NNL theory, there is a kind of irony here. For the NNL theory is predicated upon the division between the speculative order (knowledge for its own sake) and the practical order (knowledge ordered to action).[12] From the beginning, the key NNL point has been that the practical order has its own first principles that are parallel to the speculative first principles and totally underived.[13] One does not act except insofar as something is seen to be good. We only move toward something insofar as we are focusing upon its goodness. Thus, they argue continually, and with some real truth, that a good explanation of action must remain in the realm of practical truths which are all about the good that moves desire. And yet, the kind of analysis we see above is not in the practical realm, for it abstracts from an account of goodness as such. I do not incline to health according to the separation made in my mind, or knowledge cut off from life. To desire the truth presupposes the goodness of life. Even when I think it worth sacrificing my life for the truth, I am desiring in my final acts to adequate my mind to the reality, and that adequation presupposes my life. Desire does not follow the separations in thought and get cut off from what—in reality—is not separate.

And this allows us to see the broader error in the principles of the NNL theory. Consider its beginning. The NNL began as an interpretation of Aquinas’ article on natural law in the middle of his discourse on law in the Summa Theologiae.[14] The first precept of the natural law, which is also the first principle of all practical reasoning, is that good is to be done and pursued, and evil shunned.[15] This good is the end or goal of action inasmuch as in practical things the end or goal is the principle of action and thus reasoning about action.

Here is how Grisez interprets this in his foundational article. He holds that the first precept directs man to his ultimate end—which Thomas is absolutely clear about—but argues that this end is nothing definite:

The will necessarily tends to a single ultimate end, but it does not necessarily tend to any definite good as an ultimate end. We may say that the will naturally desires happiness, but this is simply to say that man cannot but desire the attainment of that good, whatever it may be, for which he is acting as an ultimate end. The desire for happiness is simply the first principle of practical reason directing human action from within the will informed by reason. Because the specific last end is not determined for him by nature, man is able to make the basic commitment which orients his entire life. The human will naturally is nondetermined precisely to the extent that the precept that good be pursued transcends reason’s direction to any of the particular goods that are possible objectives of human action.[16]

In other words, Grisez is making the same general sort of error we saw in Gorsuch and Finnis. The beginning of practical reasoning, he is arguing, is not goodness as it is in reality, but rather a non-determined ‘goodness,’ a logical concept of goodness that is open to all kinds of goods precisely because it is in fact none of them. There is a separation in thought, a separation that allows for universal predication, so that the principle of action is in things insofar as ‘good’ is predicable of them. Gone is the notion that the principle of action is the source of all goods, God Himself.[17] The NNL bases action on the conceptual good rather than the final cause. The good that grounds action is the good abstracted.

Aquinas explicitly rejects this notion of the end when he talks about how the final cause of any law is the common good. When considering the objection that law must order particular actions, Aquinas responds:

Operations indeed pertain to particular matters: but those particulars are referable to the common good, not as to a common genus or species, but as to a common final cause, according as the common good is said to be a common end.[18]

Aquinas sees that actions are not ordered to generic good, but the actual common good. Action is all about ordering our lives to the ultimate final cause. For a rational creature, this must be, finally, God Himself.[19] Grisez thinks of the good of the first precept as a mere generic orientation, not as the actual good itself. For him the ultimate end must be something constructed and based upon a vague orientation. For the NNL theorists, goodness is in the mind, not in things. Not because they think anything so silly as that we want things merely to exist in the mind, but rather because the desirability that motivates action follows the rational operations rather than preceding them.

One can see in all this the NNL’s need for the basic goods, since the ‘good’ of the first precept is insubstantial. And in the spelling out of these goods, we see the same general, logical error brought to fruition in the notion that there is no longer an intrinsic hierarchy of goods because all analysis of action and goodness comes down to the predictability that is rooted in the abstract separation of ‘good.’ Actions and things that are in reality ordered to one another are no longer so ordered, because their goodness is considered according to a separation in thought that allows for universal predication precisely because it refrains from precision.

In sum

This kind of error is by no means unique to the NNL theorists.[20] And, indeed, because the NNL theory is rightly grounded in the absolute importance of truth, its supporters avoid many erroneous conclusions about human life. Grisez for instance saw the problematic character of contraception before it was clearly understood by many.[21] And in general, because they are so focused upon the logical analysis, the NNL theorists do an admirable job, often, of breaking a situation down into its parts.[22] But they never explain the actual agency of the agent, for they do not consider goodness as such. Rather they remain in the realm of ‘things that can be called good.’

But while I suspect that most of the NNL theorists themselves would find ways to avoid the conclusions of Gorsuch, we see in his opinion the same fundamental error of logical analysis. It’s a perfect example of how this kind of reasoning—which orders desire and motivation according to the separation of concepts in the mind as opposed to the reality being analyzed by such separations—leads to absurdities in practical application. One loses sight of the real.

[1] See p. 9 of the majority opinion. This – along with the dissents of Alito and Kavanaugh – can be found at

[2] See pp. 9-10 of the majority opinion.

[3] See p. 11 of the majority opinion.

[4] See pp. 11-12 of Alito’s dissent – found in the same pdf file linked to in footnote 1.

[5] 14-15 of Alito’s dissent.

[6] One could push the original example further in other directions, though. For example, as pointed out to me by Richard Berquist, could not one make the same kind of argument that Gorsuch suggests with reference to a man (not a transgender man, to be clear) being ‘discriminated’ against because of his sex when he isn’t allowed into a women’s locker room? I presume Gorsuch would distinguish the examples, but one would have to say by the same line of reasoning that he presents in the majority opinion that the sex of the man is a ‘but-for’ condition. Take the same example and change the sex of the individual, and no one has a problem.

[7] To see a more technical account of the sort of distinctions that Gorsuch and the NNL theorists tend to miss, see Aristotle’s distinction between the subject genus and the predicate genus in his Metaphysics (Book 5, 1024b 1-20) as well as Aquinas’ commentary on these passages (Sententia Libri Metaphysicae, Lib. 5, lect. 22, no. 6-9 (Marietti). Also, in the de Ente et Essentia, Chapter 3 (Leonine ed., vol. XLIII, pp. 374-375) Aquinas details the way essence is in the logical intentions, and thus also the ways in which a logical definition is helpful but not sufficient for science. The logical intentions always treat according to a certain level of imprecision, and this why they can be predicated univocally, because the differences between things are explicitly left out. Gorsuch is bringing logical intentionality to a fight over the nature of things, you might say. (Note: you can find a nice online transcription of Aquinas’ commentary using the Marietti text as a basis here (and an English translation here), and a nice Latin and English version of the de Ente here, though you should look at numbers 50-64 inasmuch as the chapters do not line up with the Leonine edition.)

[8] Christopher Tollefsen, “The New Natural Law Theory,” Lyceum X, no. 1 (Fall 2008): 2.

[9] John Finnis, Natural Law and Natural Rights (Oxford: Oxford University Press, 2011), 92-93.

[10] The useful, the pleasant, and the honest/honorable good. See Aquinas’ Summa Theologiae (henceforth ST), I, Q. 5, A. 6, c.

[11] See in particular his De Anima, Book II, Chapter 4 (415a26-415b2) where Aristotle reflects that the preservation of the individual and the species is the way in which each material life inclines towards the divine, since it is that share in immortality and life that is possible to the material being’s activity. Rational beings have a higher participation, as one sees in the Nicomachean Ethics,Book, X, Chapter 7, (1177b26-1178a9) and the Eudamian Ethics, Book 8, Chapter 3 (1249a22-1249b25) in particular.

[12] For an introduction to this notion, one could look directly to ST, I-II, Q. 94, A. 2, c. But I suspect a more worthwhile introduction could be found in ST, I, Q. 79, A. 11, which, along with much more (such as I-II, Q. 3, A. 5), Aquinas expects the reader to have under his belt, so to speak. Practical reason is reason ordered to action, and thus practical truths are truths that are or can be the basis of the reasoning that goes into setting one’s course.

[13] Again, see Tollefsen (“The New Natural Law Theory,” 2-3) for an excellent summary of this point.

[14] ST, I-II, Q. 94, A.2, c.

[15] “Hoc est ergo primum praeceptum legis, quod bonum est faciendum et prosequendum, et malum vitandum.”

[16] Grisez, “First Principle of Practical Reason,” 199-200.

[17] See, for example, Questiones Disputatae de Veritate, 22, 2, c. (Leonine ed., v. 22, 617, 54-71). God is sought in every action implicitly because just as he is the first efficient cause that grounds all other causes, so is he the last end that grounds all other ends.

[18] I-IIae, Q. 90, A. 2, ad 2. “…operationes quidem sunt in particularibus, sed illa particularia referri possunt ad bonum commune, non quidem communitate generis vel speciei, sed communitate causae finalis, secundum quod bonum commune dicitur finis communis.”

[19] See ST, I-II, Q. 3, A. 8, c. in particular.

[20] Perhaps the single best illustration of this error is seen in the Epicureans and the Utilitarians. They confuse the pleasurable and the honest good throughout all their treatises. They implicitly see that both are delightful, but are then unable to see that there is a fundamental difference in each case with respect to where the will rests. They thus confuse the motive of action with the predication that can be common even though in reality there is a fundamental difference of motivation.

[21] See his comments to Fr. Ford (who was on the commission set up by John the XXIII and continued by Paul the VI before Humanae Vitae). In particular, note his first set of responses.

[22] Robert George, In Defense of the Natural Law, (Oxford: Oxford University Press, 1999), 46-48.