I have been intending for some time to put together a few points about natural law theory; while there are some good resources online, there are a great many that are not at all good. My goal here is primarily to lay out the basic ideas in proper order and in such a way that the most common of the many misinterpretations of these ideas can be avoided.
We must begin, then, first with that genus of which natural law theory is a species, namely, theories of practical reason. Natural law theory being the theory of practical reason qua law, we must, second, add to it the general concept of law. We must, third, say something about the features of natural law itself, of which the most important are the ways in which the precepts of natural law are known, the ordering of those precepts according to the ordering of goods, and the reasons for deviations from natural law and disagreements about its precepts. Fourth, we must say something about its relation to human law.
I. Practical Reason
Reason or rationality by its nature concerns itself with order. It is concerned with order in one way when it attempts to identify correctly the order of nature, that is, the explicit and implicit order of the world. It is concerned with order in another way when it attempts to maintain and refine in its own order, which is logical. It is concerned with order a third way when it makes decisions, which require ordering means to ends in choices and actions. And it is concerned with order a fourth way when it concerns itself with making things in appropriate ways, through building or some other skill. These four kinds of order can easily be seen to be of two kinds. The first kind is speculative or theoretical; the second kind is practical. Theoretical reasoning is concerned with correspondence between mind and what actually is; practical reasoning is concerned with correspondence between means and ends. Or, to put it another way, theoretical reasoning is concerned with the true, while practical reasoning is concerned with the good.
It is important to grasp the fact that these two kinds of reasoning are not simply separate compartments; almost all our acts, and perhaps all of them, require the confluence of both. For instance, in order to reason correctly about the world one may have to plan how to proceed in one's inquiry or to make instruments, both of which require practical reasoning; and practical reasoning in practice requires both logical thinking and reasoning about what is true of the world. Rather, what we are concerned with here is what makes the theoretical theoretical and what makes the practical practical, in all the very different interrelations they have in an actual human life.
There are surprisingly few theories of practical reason, properly speaking; bits and pieces of such theories are common, but the relatively complete theories of practical reason can arguably be counted on one hand and certainly on two: Kantianism, Mill's Art of Life and similar utiliterianisms that take the principle of utility to be not just a moral principle but a general practical principle, rationalism, and perhaps a couple of others. Natural law theory, understood as including the theory of practical reason it uses, is one of these. There are a large number of features that all such theories of practical reason share: for instance, each one insists that rationality must include some notion or account of practical rationality, each one insists that there is a reasonable sense in which people can be practically irrational, each one requires some kind of account of obligation. These are all well-motivated. We can easily recognize that there is something unreasonable about trying to discover how caterpillars become butterflies simply by staring into one's shoe, for instance, and that this is not how one ought to go about studying caterpillars. This is because we recognize that for practical acts like research there is a rational order of means to ends and an irrational order of means to ends; and that what one ought to do in order to meet the goal of your research is determined by the rational and not the irrational order of means to ends. There is, therefore, a great deal of analogy and overlap among these theories of practical reason; they tend to recognize more or less the same things as involved in reason, practical life, and practical reason itself. Where they tend to differ is in what they regard as central to these things, and this can affect, despite the many commonalities and analogies, many of the finer details. We will not be doing any comparison or contrast here, but will simply focus on natural law theory itself.
Reasoning being a kind of order, there must be some underlying principle or principles of that order. The most obvious and important of these in the case of theoretical reason is the principle of noncontradiction, which all theoretical reasoning presupposes in some way or another. For our purposes there are a few things we should note about the principle of noncontradiction. First, it is self-evident and basic: to reject it out of hand shows that you do not understand it, and it is impossible to have any reasoning at all that does not presuppose some kind of application of the principle. Second, despite this it is not a premise in every theoretical argument; it rarely is, in fact. It is instead what makes it possible for arguments to be arguments and also what draws the line, so to speak, between well developed and poorly developed arguments, whatever their premises. It structures reasoning itself. Third, and for the same reason, you can argue without ever explicitly formulating a principle of noncontradiction; being able to formulate the principle of noncontradiction in a reasonable way is very helpful for seeing the structure of reasoning, and in particular the structure of good reasoning in comparison with bad reasoning. Likewise, an erroneous formulation of the principle will not necessarily make all one's reasoning bad, although it may vitiate reasoning that explicitly uses it and may make it difficult to distinguish certain kinds of good reasoning and bad reasoning, thus making it hard for someone to correct their errors, however egregious.
Practical reasoning, however, cannot be fundamentally different from theoretical reasoning on this point; it too is a kind of order, and must have some underlying principle or principles of order. Given that practical reasoning concerns means and end, the principle or principles will have to formulate ends insofar as means can attain them. The most obvious and generally applicable of these is the one given by Aquinas, Good is to be done and sought, and bad is to be avoided. (The Latin word for 'to be done' is broad enough that it can also include 'to be made'.) Like the principle of noncontradiction, it is self-evident and basic; there is no practical reasoning that does not in some way presuppose it. A plan that has nothing good about it is a bad plan; a decision that has nothing good about it is a bad decision; whether or not things are reasonable to do or unreasonable to do, can only be determined by looking at what is good and bad about them. Like the principle of noncontradiction, the principle of practical reason is not a premise in every practical argument; it can be, but its primary role is not to be a premise but to define the character of practical reasoning and identify the most general structural difference between good practical reasoning and bad practical reasoning. Like the principle of noncontradiction, it is therefore not something that has to be explicitly formulated; it simply structures practical reasoning precisely as practical reasoning. And it occupies the same place in practical reasoning that noncontradiction occupies in theoretical reasoning. In fact, we can say something much stronger: it just is the principle of noncontradiction, considered practically. The principle of noncontradiction is drawn from the idea of being, the principle of practical reason is drawn from the idea of good, but good is just being looked at in a certain light. The principle of practical reason therefore simply identifies the form noncontradiction takes in practical matters; and, conversely, it's even difficult to talk about the principle of noncontradiction without implicitly bringing in the principle of practical reasoning above, as I did when I talked about good reasoning and bad reasoning.
There is a key difference between practical reasoning and theoretical reasoning, however, and one that will make a significant difference in how they work in practice. In theoretical reasoning you can abstract from details without affecting the reasoning itself. Since practical reasoning is precisely concerned with the details of practical life, this is not possible for practical reasoning. Practical reasoning is forced to face, all the time, a level of complexity that theoretical reasoning can often get around. This has considerable ramifications, as we shall see later.