Daily Archives: June 20, 2012

The Argument from Degrees of Perfection

The fourth of Kreeft & Tacelli’s 20 arguments for the existence of God is the Argument from Degrees of Perfection. This is also one of Aquinas’s “five ways” — the weakest of the five, in my opinion.


Summary of the argument

  1. We think of some attributes as being scalar in nature — that is, as admitting of various degrees of “more” or “less.” Examples include heat and cold, the light and dark of colors, and good and bad.
  2. Degrees of “more” and “less” imply the ideas of “most” and “least.” A continuum is defined by its two endpoints. For example, when we say one color is lighter than another, we mean that it is closer to the extreme of pure white and further from the opposite extreme of pure black. Without the extremes as standards of measurement, the idea of a continuum falls apart.
  3. Sometimes a degree of a particular attribute is communicated to an object by an outside source. For example, things are hotter when they are physically closer to a source of heat.
  4. Being itself, though it may seem like a binary quality, admits of degrees of perfection. An intelligent being exists to a more perfect degree than an unintelligent one; a being capable of love exists to a more perfect degree than one without that capacity.
  5. “But if these degrees of perfection pertain to being and being is caused in finite creatures, then there must exist a best,’ a source and real standard of all the perfections that we recognize belong to us as beings.”
  6. This perfect being is God.

There are an awful lot of things wrong with this argument, which I suspect was included in K & T’s list more for historical reasons (Aquinas!) than because anyone still finds it convincing. I mention a few of its faults below.


The idea of a continuum is prior to the idea of its extreme endpoints.

Locke spells this out with great clarity in the second book of his Essay Concerning Humane Understanding. (No, I didn’t have that reference at my fingertips. I just so happen to be reading that book at the moment.) The above argument seems to imply that we begin with the ideas of the infinitely large and the infinitesimally small, and that only with those ideas in place are we able to say that a rhinoceros is bigger than a breadbox — meaning that it is closer to the one extreme and farther from the other than a breadbox is. This is the opposite of the truth. First we notice that some finite things are larger than others, and, extrapolating from such differences, we finally arrive at the idea of immensity.


The infinite cannot serve as a standard of measurement.

Not only is it a psychological fact that we do not conceive of scalar attributes in this way; we cannot logically do so. Both the rhinoceros and the breadbox are infinitely distant from either extreme. It makes no mathematical sense to say that the rhinoceros is closer to being infinitely large than the breadbox is; therefore, that can’t be what is meant by saying that it is “bigger.”

Some continua have the form of a ray rather than line; one end terminates at some finite value, and the other extends indefinitely. For example, for temperature there is an absolute zero but no corresponding “absolute hotness.” Nevertheless, one fixed point of reference is enough. We could logically define colder as “closer to absolute zero” and hotter as “further from absolute zero.” (Psychologically, of course, that is not what we do. People who have no concept of absolute zero can understand “hotter” and “colder” well enough.)

In such ray-shaped continua, it is always the negative end of the scale which terminates in a finite and intelligible value — zero — which could be meaningfully used as a standard. “Closer to zero” means something; “closer to infinity” does not. Thus, even if we grant that degrees of perfection require some standard from which to be measured, that standard would have to be nothingness (“absolute zero”) rather than God (“absolute hotness”).


Extremes are (at best) necessary for our understanding of intermediate degrees, not for the existence of the same.

In Aquinas’s original argument (if memory serves), he stated that any being which possesses a particular attribute to some intermediate degree necessarily receives that attribute from the being which possesses it maximally — that all warm things, for example, receive that warmth from “absolute hotness” (which Aquinas identified with fire). This is such obvious baloney that K & T drop it, granting only the logically irrelevant point that sometimes hot things receive their heat from some other hot thing.

With this premise dropped, all that can be argued is that the concept of the maximum is necessary in order for us to understand the concept of a continuum — not that an actually existing maximum is necessary in order for lesser degrees of the quality in question to exist. Even if we grant the validity of this psychological point (setting aside my objections to it above), it implies nothing about the actual existence of God.


Is “being” really scalar?

Though further objections are superfluous at this point, I can’t help pointing out that the proposed scale of various degrees of perfection of being (point 4 in my summary above) seems pretty contrived, an attempt to force several qualitatively different characteristics into a quantitative continuum. Intelligence is one scale; capacity for love, another. And being itself isn’t a scale at all but a binary yes-or-no quality.

But this is not an essential point. If the argument were otherwise valid, it would still work even without this particular scale. (God would then presumably represent the standard/maximum for several different scalar attributes, such as power, wisdom, love, etc.)


Filed under God, Philosophy