What do you think?
Rate this book
The five ways: Saint Thomas Aquinas' proofs of God's existence
Philosophers and theologians who still consider belief in God to need rational justification frequently offer the arguments of Aquinas as such a justification. This book is a systematic study of the Five Ways by which St. Thomas Aquinas, in the Summa Theologiae, said that the existence of God could be demonstrated. Dr. Kenny seeks to establish what St. Thomas meant by the steps in these proofs, and he looks for illumination in parallel passages elsewhere in the works. The arguments are evaluated critically to show what they establish and what they fail to establish.
- GenresPhilosophyTheology
131 pages, Hardcover
Published January 1, 1969
About the author
Anthony Kenny
169 books122 followersSir Anthony Kenny is an English philosopher whose interests lie in the philosophy of mind, ancient and scholastic philosophy, the philosophy of Wittgenstein, and the philosophy of religion.
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 3 of 3 reviews
March 3, 2026
First Way – In this first section, Kenny clarified one of the most confusing aspects of the Five Ways, in my opinion. When reading the arguments, at seems, at first sight, that the regress Aquinas speaks of is a temporal one. Yet, that is not the case. The confusion arises from the dual meaning of the Latin verb movetur:
The causal power remains after the mover perishes, as is the case of a son still living after his parent has died. So, a series in time is not contradicting in any sense. Since Aquinas also did not see any error in a chronological causal regress, as he explicitly mentions in Summa Theologiae, I, q.46, a.2, the series is, instead, a regress simultaneous in time, in which all movers are mutually dependent.
However, despite the various other objections, the one that rises above the others centers on the Newtonian notion of inertia. Per the first law of Newtonian physics, what is at rest tends to say at rest; and what is in motion tends to stay in motion, given that an external force does not apply to this object. That said, the premise that ‘Whatever is in motion is moved by something else' is no longer true, as a mover is needed to change motion, but not to sustain it.
To conclude, the ontological regress: A is moved by B, which is moved by C, which is … ; is no longer present, because A could have been moved by B in the past, but now be moved independently through inertia, thus breaking the infinite chain.
Second Way - As in the first argument, the regress discussed in the Second Way is not a chronological one. Instead, it stretches into the heavens simultaneously. The argument is based on the medieval astrological model of causation, in which the heavens (celestial bodies) directly cause changes on earth in a hierarchical chain of simultaneous causes. In this model, God moves the celestial spheres, which in turn influence events on Earth, all in a simultaneous manner, such that all motion ultimately depends on the perfect, eternal motion of celestial bodies. Kenny states:
The equivocation Kenny mentions is the seemingly ambiguous use of the term first, which is ultimately used as first in ontological priority, but heavily implies first in time when considering the impossibility of an infinite series. That is certainly the case in my personal experience, as I initially took his meaning in a chronological sense.
From this, we understand that the First and Second Ways are really similar since both acknowledge a hierarchical (simultaneous) regress series, both admit this series cannot regress infinitely, and both propose, as a solution, a First Cause to break this chain. Thus, the question becomes wether this hierarchical regress is impossible or not. While I’m not fully convinced by Kenny’s critique, as to me the regress of the Second Way goes deeper than a quick cosmological dismiss can counter, I still reject the argument. Why posit a first principle at all? An investigation in the ultimate explanation of the universe is fundamental to our understanding of reality as a whole, yet while science does not arrive at a sufficiently plausible explanation (stronger than the Big Bang), dealing in a metaphysical question such is this is not going to yield definite results. It’s not that discussion is useless, but the fundamental way we live and deal with reality does not involve the ultimate first cause or the ontological regress at all. Reality is; independent of its metaphysical basis.
Third Way - The argument from contingency starts by differentiating the definition of necessity that Aquinas uses. Avicenna’ necessity is a logical one, in which a necessary being is one that its non existence results in a contradiction. In contrast, Aquinas defined necessity in terms of unalterability, so a necessary being is one that cannot cease to exist once it exists. That is due to his innovation involving God’s essence, which is the same as His existence. Thus, he rejects Avicenna’s definition because that implies the use of existence of a predicate, which goes against the Divine Simplicity.
Kenny’s critique focuses on rejecting the premises of the argument. Premise (1), for instance, isn’t true. Possibility does not entail actuality; a possibly non-existent being does not mean that in a point in time it ceased to exist. As an example, the universe might be logically possible to not exist, yet it never ceased to do so.
Nevertheless, the greatest flaw in the argument lies in premise (2), which even in my first read was obvious to me. Implicitly, it states that if each thing at some point in time is not, then at some point in time nothing is. Formally speaking, Kenny identifies that as the quantifier-shift fallacy. Granting (1) for the sake of the argument, every contingent thing could fail to exist at some time, but that does not mean that every contingent thing fails to do so at once. It is entirely coherent to imagine a perpetual succession of contingent beings such that at every point in time, at least one exists.
Moreover, the argument presupposes that this point in time when nothing exists is in the past. This is, of course, not necessary, as the point in time when all contingent beings cease to exist could very well be in the future. However, the arguments needs this moment to be in the past, as premises (3) and (4) imply.
Fourth Way - Kenny draws the parallels between the Fourth Way and the Platonic theory of Ideas. The argument resembles this theory remarkably, but instead of particulars participating in the Idea, they resemble or approach that which is the ultimate standard, according to Aquinas.
However, that is not the case. Particulars do not participate in metaphysically real Ideas in the Platonic sense; they also don’t resemble God’s perfections in the analogous sense, as Aquinas posits. That is because this standard does not exist in the first place. When using these predicates, are the degrees measured against a fixed, unchanging maximum? For instance, something hot is only hot as a comparison to something colder. “Hotness” is not a universal quality, rather it is relative to what is hotter than our skin. Additionally, what is good may be good in different senses, depending on the context. Is goodness in these various uses measured against the same maxim? Or are there various standards of “goodness”, depending on the context? That said, I question if the argument moves from comparative-based language to metaphysical conclusions without justification.
In contrast, Kenny’s critique goes deeper than only the Fourth Way, challenging much of Aquinas metaphysics. It lies in Aquinas’ treatment of Being as something close to a platonic form. From observing many things that have being, he moves to a universal concept of Being which functions as a ‘Form’ of Being. This predicate may have two meanings: firstly, it is a normal predicate that applies to everything. If that is the case, the predicate itself does not add any information to what is predicated of it, such that “Socrates exists” would be the same as “Socrates”. Then, Being is a predicate too uninformative to be relevant. The other interpretation is that Being is not a regular predicate. In this view, “Socrates exists” entails that there is such a thing as Socrates, not that Socrates has the property of existing. Extrapolating a predicate as a thing, treating it as something which can itself be that predicate is wrong in the case of Being. For instance, ‘redness’, when treated as a thing, may be “a red thing”; ‘tallness’, with the same treatment, may be “a tall thing”. However, Being becomes “an existing thing”, which is absurd since this second interpretation admits Being as something which cannot be predicated in a normal sense, which is not the case in “an existing thing”. Kenny’s argument, in summary, reads as:
Fifth Way - Aquinas argues from the regularity of nature. Nature, more specifically inanimate nature, is consistent, producing regular outcomes that act as a goal. Since things that lack awareness cannot aim at an end unless directed by something intelligent, they must be intelligently directed in order to reach these regular outcomes (goals) reliably.
But why must goal-directed outcomes require intelligence? The regularity of nature, which is the principle that enables science, could very well be explained by mechanistic laws, as is the case today. Nature is explained through chemical structure, causal necessity, physical laws, etc; which does not presupposes an orchestrating intelligence.
Additionally, why should nature be teleologically geared at all? Artificial objects, indeed, are always made with a goal in mind: a knife to cut; a light bulb to illuminate an environment; a clock to track time. But natural instances do not intrinsically require a goal. Kenny ably showcases this:
Therefore, the Fifth Way fails when it it admits that regularity necessitates a teleological gearing, which in turn necessitates an intelligent being to direct nature.
--------
To conclude, I read this work to deepen my understanding of both the Five Ways and why they ultimately fail. This was an opportunity to better evaluate the necessity of God from a rational/logical perspective, since in my essay on religion I skipped a discussion of God’s existence in favor of a narrower focus. That said, the arguments obviously fail to sway both Kenny and I, which does not come as a surprise even though I’m always open for change. His objections are sharp, even though sometimes they come across as a little cryptic, especially towards the last two arguments. Nevertheless, it was a valuable experience, in which I certainly learned a lot about not only the philosophy of religion, but also of how arguments can be structurally analyzed, and the nature of metaphysical discussions.
Whatever is in motion is moved by something else'. Does this mean that whatever is now in motion is now being moved by something else? Or does it mean that whatever is in motion has, at some time or other, been moved by something else? That is to say, how seriously are we to take the present tense of the verb movetur, which can mean either 'is being moved' or 'is (at some time or other) moved'? If we take the former interpretation, then the series of moved movers which Aquinas goes on to discuss will be a series of simultaneous movers; if the latter, then it may be a series stretching backwards in time.
The causal power remains after the mover perishes, as is the case of a son still living after his parent has died. So, a series in time is not contradicting in any sense. Since Aquinas also did not see any error in a chronological causal regress, as he explicitly mentions in Summa Theologiae, I, q.46, a.2, the series is, instead, a regress simultaneous in time, in which all movers are mutually dependent.
However, despite the various other objections, the one that rises above the others centers on the Newtonian notion of inertia. Per the first law of Newtonian physics, what is at rest tends to say at rest; and what is in motion tends to stay in motion, given that an external force does not apply to this object. That said, the premise that ‘Whatever is in motion is moved by something else' is no longer true, as a mover is needed to change motion, but not to sustain it.
To conclude, the ontological regress: A is moved by B, which is moved by C, which is … ; is no longer present, because A could have been moved by B in the past, but now be moved independently through inertia, thus breaking the infinite chain.
Second Way - As in the first argument, the regress discussed in the Second Way is not a chronological one. Instead, it stretches into the heavens simultaneously. The argument is based on the medieval astrological model of causation, in which the heavens (celestial bodies) directly cause changes on earth in a hierarchical chain of simultaneous causes. In this model, God moves the celestial spheres, which in turn influence events on Earth, all in a simultaneous manner, such that all motion ultimately depends on the perfect, eternal motion of celestial bodies. Kenny states:
It is by now clear that the series of causes from which the Second Way starts is a series whose existence is vouched for only by medieval astrology. The Second Way, like the First, starts from a series of simultaneous agents and not from a series stretching backwards in time; the Second Way, like the First, uses an equivocation between 'first= earlier' and 'first= unpreceded' to show that this series cannot be an infinite one. But the First Way starts from an indisputable fact about the world; the Second starts from an archaic fiction.
The equivocation Kenny mentions is the seemingly ambiguous use of the term first, which is ultimately used as first in ontological priority, but heavily implies first in time when considering the impossibility of an infinite series. That is certainly the case in my personal experience, as I initially took his meaning in a chronological sense.
From this, we understand that the First and Second Ways are really similar since both acknowledge a hierarchical (simultaneous) regress series, both admit this series cannot regress infinitely, and both propose, as a solution, a First Cause to break this chain. Thus, the question becomes wether this hierarchical regress is impossible or not. While I’m not fully convinced by Kenny’s critique, as to me the regress of the Second Way goes deeper than a quick cosmological dismiss can counter, I still reject the argument. Why posit a first principle at all? An investigation in the ultimate explanation of the universe is fundamental to our understanding of reality as a whole, yet while science does not arrive at a sufficiently plausible explanation (stronger than the Big Bang), dealing in a metaphysical question such is this is not going to yield definite results. It’s not that discussion is useless, but the fundamental way we live and deal with reality does not involve the ultimate first cause or the ontological regress at all. Reality is; independent of its metaphysical basis.
Third Way - The argument from contingency starts by differentiating the definition of necessity that Aquinas uses. Avicenna’ necessity is a logical one, in which a necessary being is one that its non existence results in a contradiction. In contrast, Aquinas defined necessity in terms of unalterability, so a necessary being is one that cannot cease to exist once it exists. That is due to his innovation involving God’s essence, which is the same as His existence. Thus, he rejects Avicenna’s definition because that implies the use of existence of a predicate, which goes against the Divine Simplicity.
(1) Whatever has the possibility of not being, at some time or other is not.
(2) If everything has the possibility of not being, then at some time or other nothing is.
(3) If at some time or other nothing is, then there is nothing now
(4) It is not the case that there is nothing now.
(5) Not everything has the possibility of not being.
Kenny’s critique focuses on rejecting the premises of the argument. Premise (1), for instance, isn’t true. Possibility does not entail actuality; a possibly non-existent being does not mean that in a point in time it ceased to exist. As an example, the universe might be logically possible to not exist, yet it never ceased to do so.
Nevertheless, the greatest flaw in the argument lies in premise (2), which even in my first read was obvious to me. Implicitly, it states that if each thing at some point in time is not, then at some point in time nothing is. Formally speaking, Kenny identifies that as the quantifier-shift fallacy. Granting (1) for the sake of the argument, every contingent thing could fail to exist at some time, but that does not mean that every contingent thing fails to do so at once. It is entirely coherent to imagine a perpetual succession of contingent beings such that at every point in time, at least one exists.
Moreover, the argument presupposes that this point in time when nothing exists is in the past. This is, of course, not necessary, as the point in time when all contingent beings cease to exist could very well be in the future. However, the arguments needs this moment to be in the past, as premises (3) and (4) imply.
Fourth Way - Kenny draws the parallels between the Fourth Way and the Platonic theory of Ideas. The argument resembles this theory remarkably, but instead of particulars participating in the Idea, they resemble or approach that which is the ultimate standard, according to Aquinas.
However, that is not the case. Particulars do not participate in metaphysically real Ideas in the Platonic sense; they also don’t resemble God’s perfections in the analogous sense, as Aquinas posits. That is because this standard does not exist in the first place. When using these predicates, are the degrees measured against a fixed, unchanging maximum? For instance, something hot is only hot as a comparison to something colder. “Hotness” is not a universal quality, rather it is relative to what is hotter than our skin. Additionally, what is good may be good in different senses, depending on the context. Is goodness in these various uses measured against the same maxim? Or are there various standards of “goodness”, depending on the context? That said, I question if the argument moves from comparative-based language to metaphysical conclusions without justification.
In contrast, Kenny’s critique goes deeper than only the Fourth Way, challenging much of Aquinas metaphysics. It lies in Aquinas’ treatment of Being as something close to a platonic form. From observing many things that have being, he moves to a universal concept of Being which functions as a ‘Form’ of Being. This predicate may have two meanings: firstly, it is a normal predicate that applies to everything. If that is the case, the predicate itself does not add any information to what is predicated of it, such that “Socrates exists” would be the same as “Socrates”. Then, Being is a predicate too uninformative to be relevant. The other interpretation is that Being is not a regular predicate. In this view, “Socrates exists” entails that there is such a thing as Socrates, not that Socrates has the property of existing. Extrapolating a predicate as a thing, treating it as something which can itself be that predicate is wrong in the case of Being. For instance, ‘redness’, when treated as a thing, may be “a red thing”; ‘tallness’, with the same treatment, may be “a tall thing”. However, Being becomes “an existing thing”, which is absurd since this second interpretation admits Being as something which cannot be predicated in a normal sense, which is not the case in “an existing thing”. Kenny’s argument, in summary, reads as:
The objections, therefore, which have been made during the centuries to Plato's theory of Ideas apply to Aquinas' theory of God as subsistent being. But this is not all that is wrong with the theory: for the predicate to which the Platonic extrapolation is applied is, as we have seen, a very extraordinary predicate. Either it is understood as a predicate which holds of all substances; in which case it is too uninformative to constitute the essence of any entity ; or it is understood as a variable expression which permits of no substitution, in which case it is an ill-formed formula. The notion of Ipsum Esse Subsistens, therefore, so far from being a profound metaphysical analysis of the divine nature, turns out to be the Platonic Idea of a predicate which is at best uninformative and at worst unintelligible.
Fifth Way - Aquinas argues from the regularity of nature. Nature, more specifically inanimate nature, is consistent, producing regular outcomes that act as a goal. Since things that lack awareness cannot aim at an end unless directed by something intelligent, they must be intelligently directed in order to reach these regular outcomes (goals) reliably.
But why must goal-directed outcomes require intelligence? The regularity of nature, which is the principle that enables science, could very well be explained by mechanistic laws, as is the case today. Nature is explained through chemical structure, causal necessity, physical laws, etc; which does not presupposes an orchestrating intelligence.
Additionally, why should nature be teleologically geared at all? Artificial objects, indeed, are always made with a goal in mind: a knife to cut; a light bulb to illuminate an environment; a clock to track time. But natural instances do not intrinsically require a goal. Kenny ably showcases this:
Water, when cooled, turns into ice and not into, say, coal. But what has this to do with goals? Explain, if you like, the freezing of water by saying that water has a natural tendency to freeze; but why say that water is aiming at anything in freezing? Aquinas' argument suggests that what is being aimed at is the freezing itself; but this amounts to saying that water freezes because it freezes.
Therefore, the Fifth Way fails when it it admits that regularity necessitates a teleological gearing, which in turn necessitates an intelligent being to direct nature.
--------
To conclude, I read this work to deepen my understanding of both the Five Ways and why they ultimately fail. This was an opportunity to better evaluate the necessity of God from a rational/logical perspective, since in my essay on religion I skipped a discussion of God’s existence in favor of a narrower focus. That said, the arguments obviously fail to sway both Kenny and I, which does not come as a surprise even though I’m always open for change. His objections are sharp, even though sometimes they come across as a little cryptic, especially towards the last two arguments. Nevertheless, it was a valuable experience, in which I certainly learned a lot about not only the philosophy of religion, but also of how arguments can be structurally analyzed, and the nature of metaphysical discussions.
June 30, 2026
A fairly brief and yet detailed exploration of Aquinas' 5 Ways. Kenny exposition upon them presupposes some prior understanding, owing to the generally scathing tone he presumes throughout. He is especially good in detailing the equivocations present within the Latin texts, showing the interpretative implications that follow from differing translations. This is especially present in his discussion regarding the 1st Way, the Argument Ex Motu. As far as I am aware, Kenny was one of the first in Analytical Philosophy of Religion to properly raise the objection from what is now dubbed 'Existential Inertia'. He also provides a good critique of the 3rd Way, or Argument from Contingency, detailing some of the semi-fallacious errors present within the deductions, most clearly uncovered by formalizing the argument into FOL. The 4th and 5th Way's are also considered in good detail. Perhaps my only annoyance was I felt that the 2nd Way, the argument from Efficient Causality, could have been covered in a slightly better way.
January 30, 2022
This book is highly impactful.
Displaying 1 - 3 of 3 reviews