Dialectic as a Way to the Principles


Dr. Anthony Andres
Tutor Talk
Thomas Aquinas College, California
January 22, 2020


In the second chapter of the Topics, his treatise on dialectical reasoning, Aristotle tells us that dialectic is useful for three things: intellectual exercise, conversation, and the philosophical sciences. He then divides the third use into two. He writes:

[It is useful] for the philosophical sciences because, when we are able to raise doubts on both sides, we easily perceive the true and false in each. Moreover, [it is useful] for the first principles of each science. For nothing is able to be said about these [first principles] from the proper principles of the proposed science, since they are the first of all principles; but it is necessary to speak about these from those things which are probable in each. This, however, is either proper, or at least most fitting, to dialectic. For, since [dialectic] is able to examine {test}, it contains the way to the principles of all the sciences.

The truth of these assertions is clear from our own experience here: most of our classes consist primarily of the continual dialectical probing which seems necessary to make clear to ourselves, not only the conclusions that we learn, but also the premises by which we learn them. Perhaps that experience ought to be enough for me, but in fact I have for many years asked myself how dialectical reasoning helps us to grasp the first principles of the philosophical sciences. I hope that this paper will be a beginning for understanding the answer to that question.

The subject of this paper, then, is the relation between dialectic and the first principles of the philosophical sciences, and the question that it is trying to answer is why dialectic is a road to grasping those principles. I will argue in my paper that dialectic is such a road because dialectical reasoning about the principles leads us from confusion to distinctness in our grasp of the subject term of those principles. My paper falls into three parts: in the first I take up what dialectical reasoning is, showing how it differs from demonstration; in the second part I will discuss the nature and properties of the first principles of the demonstrative sciences; and in the third I will try to explain how a dialectical consideration of a first principle makes it known.

‘Dialectic’ is a word with many meanings in philosophy, but we will look only at the ways in which Aristotle uses the word in the Topics. It principally names a faculty for making a dialectical syllogism. In both the Prior Analytics and the Topics Aristotle defines the dialectically syllogism by contrasting it to the demonstrative syllogism. As we learn in the Posterior Analytics, the demonstrative syllogism is reasoning that begins from premises that are true and primary and that yields a conclusion known with certainty. The propositions of Euclid are the kind of things that Aristotle had in mind when he defined demonstration in that way. Euclid begins his science by laying out definitions, postulates and common notions which he assumes that we know with certainty, and in the propositions that follow he reasons to conclusions that necessarily follow from these.

In contrast, the dialectical syllogism or induction does not take for premises what is true and primary, but rather what is probable. Of course, this does not mean that dialectical premises are necessarily false; it means instead that dialectic does not assume that its premises are known with certainty. What it does assume is that the premises are probable, that is, generally approved, either by all or most men, or by the wise. Examples of premises probable in this way are “Friends show each other signs of affection” and “Love and hate dwell in the same part of the soul.” These statements garner the qualified assent of most men because they fit with the very ways in which we talk about their terms. For example, we talk about love and hate as opposites, qualities that cannot be present in the soul at the same time. If they cannot be present together, it must be because they inhabit the same part of the soul. When love enters, hate is driven out. And so the dialectical syllogism begins from premises that are not certain, but only probable, and probable because they fit the ways that men talk about their terms.

So, because demonstration begins with premises that are true and certain, one demonstrative argument by itself is capable of being fully convincing to the learner. But because a dialectical syllogism begins from premises that are merely probable, it does not produce much certainty about its conclusion. In order to increase the certainty of his conclusion, the dialectician will offer as many arguments as possible in its support. And so it is a feature of dialectic that it offers many arguments through many middle terms in order to produce as much conviction as it can about its conclusion. This feature of dialectic will become important later in our inquiry.

After Aristotle distinguishes the dialectical from the demonstrative syllogism, he discusses the usefulness of this treatise on dialectical reasoning. As we saw before, he gives us three uses for dialectic, intellectual exercise, conversation and philosophical science. St. Albert the Great, in his commentary on the Topics, reduces the first two to the third. He argues that we study the treatise, the Topics, so that we can exercise our ability to reason about philosophical questions from probable premises. And we then use that ability in conversations in order to remove the obstacles in the minds of others to the correct understanding of philosophical truth. So, according to St. Albert, the other two uses find their ultimate rationale in the use of dialectical reasoning in the philosophical sciences.

As we saw before, one use in the philosophical sciences is to consider arguments both for and against a conclusion in such a science. For example, Aristotle gives arguments both for and against the existence of the void in Book IV of his Physics before he attempts to reach a definitive conclusion. But our concern here is why he thinks that dialectic is useful for grasping the principles of all the sciences.

Aristotle’s brief account has two parts. In the first part he explains why we cannot discuss the principles scientifically, and here it makes sense to recall our previous example, Euclid’s Elements. The propositions in the Elements are the conclusions to demonstrative arguments; since they are conclusions, our knowledge of them relies on our knowledge of the premises from which they are concluded. Some of those premises are themselves the conclusions of previous propositions, but ultimately everything depends upon the first premises, the definitions, postulates and common notions which he outlines at the beginning of the first book. These are the first truths of the geometry, and no premises are prior to them. Therefore, we cannot use geometrical premises to argue for or against the definitions, postulates or common notions. And Aristotle concludes universally that it is impossible to prove or discuss the principles of a science from the premises that are proper to that science.

In the second part of his account Aristotle argues that we can use dialectic to reason about the first principles of the philosophical sciences. Although we cannot argue about the principles of any science from what is proper to that science, he points out that we can argue about them from what is probable, that is, from what is generally approved about them. For example, we cannot argue geometrically that all right angles are equal because that statement is a first principle in geometry. But if we need to, we can discuss the question from probable statements about angles. But to reason from what is probable is the function of dialectic; therefore, a treatise on dialectic such as the Topics will be very useful for the pursuit of the philosophical sciences.

But at this point a problem always arises in my mind. The first principles are first principles and so cannot be proven by a demonstration. They are, as St. Thomas puts it, per se nota, known through themselves, not through being deduced from some prior premises. Moreover, since they are the source of our knowledge of the conclusion, they are actually more known than any of the conclusion that follows from them. They are known to be true with the greatest certainty. But the premises of dialectical syllogisms are not certain but merely probable, and their conclusions, like the conclusions of all syllogisms, are no more certain than their premises. And even the combination of many dialectical arguments does not produce the certainty about the conclusion that must belong to a first principle. In sum, because the first principles are per se nota, they seem not to need any support that would be provided by reasoning to them; and because dialectical reasoning begins from mere probabilities, it seems unable to provide the necessary support. It seems, then, that dialectical reasoning would be entirely useless for our grasp of the first principles of demonstration.

This problem brings into focus what we need to look at in the next part of our inquiry: we need to understand what a first principle is, what it means for something to be per se nota, and we especially need to see how we come to know these first principles. We will use texts from both Aristotle and St. Thomas to help us answer these questions. We will first look at the principles from a logical point of view, then from the point of view of the higher sciences.

Aristotle discusses the first principles at the beginning and end of his treatise on demonstration, the Posterior Analytics. In the second chapter of the first book he assigns to them five characteristics: they must be “true, first, immediate, better known than and prior to the conclusion, and they need to be related to conclusion as effect to cause.” He points out that they must be true, and known to be true, because although we can reach a true conclusion from false premises, we cannot know a conclusion to be true from false premises. He points out that the premises must be first in the sense of being indemonstrable, since if they were demonstrable they would require demonstration to be known, and thus there would be premises prior to them. And he tells us that, because they are indemonstrable, they are immediate; that is, they are not known through some middle term. Finally, because science is knowledge of something through its cause, he explains that they must be the cause of the truth of the conclusion, prior to and better known than it. These, then are the logical characteristics of the first principles.

St. Thomas, looking at these same principles from the perspective of higher sciences, assigns a different set of characteristics to them. In various places he notes that the first principles are known naturally, that they are conceded by all, that they are known as soon as their terms are known, and that they are so certain that no one is able to think the opposite of them. But the characteristic that he most often assigns to them, and which seems to be the root of the others, is that the predicate of a first principle is in, or from, the very ratio of its subject. So our next task should be to consider carefully what St. Thomas means when he says this.

St. Thomas tells us that the principles naturally known are infused in the minds of men through the light of the agent intellect. I think we should understand this as working in the following way. The agent intellect makes a potentially intelligible nature, contained in the sensible image, actually intelligible and understood, present in the intellect which becomes all things. When another nature which is immediately rooted in the first is also understood, and when it is compared to the first nature, by the light of the agent intellect we see the second nature as coming from the first nature and as in some way being present in the first nature. And we see this without having to use some third nature as a middle connecting the two. And since we see the second nature as being present in the very nature of the first, we cannot help affirming the statement which joins the two natures as soon as we understand the terms of that statement.

The example that St. Thomas often gives of such a principle is the axiom that the whole is greater than its part. He points out that to be greater than its part is something that is contained in and flows from the nature of a whole; the intellect seeing the nature of ‘whole’, sees that being greater than the part is immediately in and from that nature, and so the intellect cannot help but predicate the former of the latter. And consequently the proposition that the whole is greater than the part is known naturally, is conceded by all, is known as soon as the terms are known, and is known with such certainty that no one is able to think the opposite. And so in general the fact that the nature of the predicate is in some way present in the nature of the subject is at the root of all of these features of the self-evident principles of knowledge.

In another place St. Thomas points out that the principles are made know to us immediately through sense, as opposed to being made known through a rational investigation. This fits, of course, the account that we have given of their being made known by the light of the agent intellect, but it also brings to mind Aristotle’s discussion at the end of the Posterior Analytics of how we come to know the first principles of demonstration. Since the Posterior Analytics teaches logic, it does not make sense for Aristotle to invoke there the power of the agent intellect. But even the logician can see three things about the process of coming to know the first principles. First, he can see that man is not born with knowledge of the first principles; we could not have a most excellent form of knowledge without being aware that we have it. Second, he can see that, since all learning comes from pre-existent knowledge, and the only knowledge prior to that of the first principles is sensation, then our knowledge of the first principles must be derived in some way from sensation. Third, because what is sensed is individual, but what is understood is universal, he can also see that the process will have something in common with induction.

Of course, the problem that Aristotle cannot entirely solve in the Posterior Analytics is how an inferior form of knowledge, sensation, can be the origin of a more excellent form, understanding, but his account does helps us see from another angle what it means for the first principles to be naturally known. It does not mean that the soul possesses knowledge of the first principles from birth or conception; Aristotle is clear that we learn the first principles. What it rather seems to mean is that the natural activity of the intellect, in contrast to the deliberate activity of rational investigation, is enough to bring about in our minds knowledge of these principles, knowledge so certain that we are not able to even think the opposite.

This conclusion, however, seems only to sharpen the problem about the use of dialectic in the philosophical sciences. If the first principles are naturally known by the light of the agent intellect without the investigation of reason, it seems that we cannot help coming to know the first principles. And if we cannot fail to learn the first principles, it seems that dialectic, a deliberate method of rational investigation, would be superfluous for learning them. Dialectic, then, still does not seem to be a way to the principles.

But a distinction that St. Thomas and Boethius make between two kinds of principles helps us solve this problem. St. Thomas divides the per se nota principles of the sciences into common conceptions of the soul, which are known to all, and proper principles, which are known only to the wise. What is the difference? As we saw before, everything per se nota is known as soon as the terms are known because the predicate is found in the very ratio of the subject. Some terms, such as whole and part, are known to all men, and so the per se nota statement that every whole is greater than its part is also known to all; but other terms, the terms proper to the particular sciences, are not known to all, but only to those with an experience of the science. The term “right angle” is an example. That term is proper to the science of geometry, and so not everyone understands what a right angle is, but only those who have some experience of geometry. Therefore, although the proposition “Every right angle is equal” is per se nota, it is not known to all men, but only to those who understand the nature of its subject, the right angle.

What we especially need to take away from this distinction is that, in some cases, it is possible for a statement to be per se nota in itself but still to be unknown to some men. We also see why this happens: some men fail to grasp the subject of a per se nota statement. By itself these assertions might make us think that there is room here for dialectical reasoning to help us learn the truth of at least some first principles of science. Perhaps, then, we have already answered our original question.

But in fact we still run into a problem. On the one hand, if we know the subject of the per se nota statement, then we can immediately grasp its truth, and so a dialectical argument supporting it seems superfluous. On the other hand, if we do not know the subject, we cannot consider the statement at all, and so a dialectical argument supporting it seems impossible. In either case it seems that dialectic still cannot help us understand the principles.

A text that can help us solve this new problem, and perhaps the whole problem, is found at the beginning of Aristotle’s Physics. There Aristotle concludes that natural science should take up first the consideration of the most universal principles of nature. He writes:

It is necessary to proceed in this way, from what is less certain by nature but more certain to us, toward what is more certain and more knowable by nature. But the things which are first obvious and certain to us are rather confused, and from these the elements and principles become known later by dividing them. Whence, it is necessary to go from the universal to the particulars.

What is important about the passage above is the reason that we need to go from the study of universal principles to that of particular ones: the confused is more known to us than the distinct. Aristotle manifests this in two ways, through sensation and through intellectual knowledge. He points out that what is first sensed is a confused whole, and only after we have perceived that do we come to a distinct sensation of the parts. St. Thomas gives the example of seeing, first the whole house, and then its parts.  But Aristotle points out that the same progression also happens at the level of intellect: a name indistinctly signifies a whole nature, while the definition “divides into single parts,” that is, distinguishes the principles of that nature. We have to name something before we can state its definition because we understand the whole nature confusedly first, and only after this distinctly understand the principles of that nature. So both in sensation and in understanding we know something first in a confused way, and then in a distinct way.

And so, St. Thomas points out in his commentary, between not knowing a subject and knowing that subject distinctly there is a third possibility, knowing that subject in a confused way. And this means knowing something of the nature of the subject, but not having a distinct grasp of the principles of its nature. This enables us to name the subject, assign predicates to it, and even to make it the term of a syllogism, but it does not enable us to see the truth of a self-evident statement about it with certainty.

Even when our knowledge of the nature is confused, however, the principles of the nature of that subject are still in some way present in our conception it, perhaps implicitly or virtually. But, and here is the key point, they are not present in such a way that we can draw out of them the predicate which would immediately flow from them. For example, we might know enough about the right angle to name it, but not enough to see that equality to all right angles flows from the very ratio of the subject. We might be caught in the middle, knowing enough about the right angle to understand what the self-evident statement means, but not enough to see that it is immediately true.

I think that another example of this would be the first definition of the soul, substance as species of a natural body with life potentially. Before we undertake the study of the soul, we know enough about the soul to name it and know that it exists, but we do not see its principles distinctly. And thus, we do not know enough about the subject of the following statement, “The soul is substance as species of a natural body with life potentially,” to see that this statement is immediately true. And so before we can come to see the truth of this statement, we will have to study the soul and reason about it from the opinions that all men or wise men share about the soul. That is, the definition of the soul will be made clear through dialectical reasoning.

It looks, then, like we have solved all of our problems. We now see that it is not impossible, at least in some instances, to be ignorant of first principles, statements that are self-evidently true. We are ignorant when we do not know the nature of the subject of the statement. Also, we have seen that knowing or not knowing the principle is not just a question of knowing its subject perfectly or not knowing it at all. Rather, there is a middle state between the two, knowing the subject in a confused way. In this case, we know enough about the subject to be able to name it and to discuss the self-evident statement about it, but not enough to see immediately that the principle is true. All of this leaves room for the operation of dialectic in our grasp of the principles.

Our final task is to explain how dialectical arguments move our minds from a confused grasp of the subject to a distinct one. Two factors are present in this account. First, St. Albert points out that the probable statements which we use in dialectical reasoning have the character of signs. I think this means that we are naturally inclined to accept these statements because the principles of their subjects are implicitly present in our conception of them. For example, even before we have a definition of the soul, being a substance is in some imperfect way present even in our confused grasp of the soul. This vague grasp inclines us to deny that one living thing has many souls, and that denial is a sign of the presence of the principles of substance in our vague grasp of the nature of the soul. So our natural assent to the probable statement is a sign of the presence of some principle in the nature of its subject.

The second factor is one mentioned much earlier: dialectic proposes many different arguments about the same subject and in support of the same statement. The arguments are many because the probable statements from which they proceed are many and distinct. I remember  one morning several years ago discussing with another tutor some principle in mathematics. That afternoon I received an email with 18 arguments against my position. That was a remarkable display of the power of dialectic.

If we combine the two factors, that the probable statement is a sign and that dialectic produces many arguments, we see how dialectic can lead to the grasp of a first principle of a science. The subject of the first principle is also the subject of many distinct dialectical arguments and so of many distinct probable premises. Each of those premises is a sign of the imperfect presence of some principle of the subject in our conception of it. Therefore, the many distinct probable statements can be signs of many distinct principles in the essence of the subject. But once we see distinctly the principles in the subject, we see that the predicate is already present in the subject in some way, and we immediately predicate it of that subject and affirm the first principle of the science. Thus, the multiplicity of dialectical arguments is an important aid in the grasp of the proper first principles of the sciences, at least those known only to the wise. Dialect, then, truly is a way to the principles of all sciences.

Our inquiry into this question has been long and difficult, but I hope that it has at least made us more aware of the nature of the first principles of the sciences and of what it means to be per se nota or self-evident. I also hope that it has made us aware of how difficult it can sometimes be to grasp those principles and how the dialectical method that we use at this college is a good way to reach them.

Finally, I want to point out that the use of dialectic in grasping the principles might go further than I have discussed. In this paper I have only explored how it might be used to grasp the principles proper to a science; I have left aside the consideration of the common principles, the ones that everyone knows in some way. But I think that there is room for the use of dialectic even in these cases. After all, the common first principles are also subject to confusion, though of a different kind than the proper principles. It is not so much the confusion of the principles of their subject, but the confusion of the statement as a whole with the particulars with which it is almost invariably considered.  But I have already taken up too much time, and I put this forward as a suggestion, not as a worked out thesis.


Streaming & downloadable audio


Receive lectures and talks via podcast!


Get it on iTunes
Available on Stitcher