What Is Ockham’s Razor?

William of Ockham—philosopher, logician, theologian, Franciscan friar, and enemy of the church—has largely been accredited with the idea of Ockham’s razor. Often used within the philosophy of science, it has gained fame and notoriety since its conception in the Middle Ages. Ockham’s razor is a principle that prefers elegance, parsimony, and leaving behind ontological baggage. So, who was William of Ockham, and how sharp is his infamous razor?

The idea is largely attributed to—and named after—William of Ockham (c. 1287–1347). He was one of the key figures of the High Middle Ages, along with other big scholastic names such as Thomas Aquinas or John Duns Scotus. However, unlike the others, Ockham was more of a rebel. He was charged with heresy by the Papal Court at Avignon and later excommunicated for fleeing Avignon.

Besides his political writings, which were controversial for that period, Ockham’s work in logic and metaphysics was equally unorthodox because he was a nominalist.

This means that he believed there are no metaphysical universals—such as Platonic Forms—and that ontology should be stripped down to its fundamental categories. His nominalism was underpinned by the principle we now call Ockham’s Razor.

The debate about universals was one of the key philosophical and theological debates that shaped the Middle Ages. The whole fuss was about the existence of universals – are they real and mind-independent entities with similar characteristics or mere words used for categorizing particular things? Philosophers and theologians inclined to the former view were called realists and followed in the footsteps of Plato, while nominalists opted for the latter view.

A Crash Course on Universals

So, why was everyone in Medieval philosophy in such a huff about the existence (or lack thereof) of universals?

The debate had important theological implications about the nature of God and the status of moral truths. If we wanted to claim that God has abstract qualities such as goodness and that our good deeds represent only instantiations of such universal goodness, then realists would wholeheartedly approve our claim.

Nonetheless, nominalists would be pretty disappointed with this (and us, for that matter). For them, abstract qualities can’t be independent of our minds. They are umbrella terms. Thus, our good deeds may be called “goodness,” this term may apply to God, but it is ultimately flatus vocis, or “breath of voice.” This could mean that universal moral truths can’t be derived from God and have no divine justification but are relative to different societies and periods.

In the Middle Ages, these were unacceptable and dangerous lines of argumentation. One could, God forbid, go further and say that the Church doesn’t gather all believers yearning for salvation in a single spiritual congregation close to God but is merely one religious community out of many. You don’t want to know how this could escalate back then.

The point is, however, that Ockham’s nominalism would be suspicious from the very start, even if he hadn’t developed it further in his political writings. For him, God was the only absolutely necessary entity, and he did not deny the existence of abstract qualities. So far, so good!

Unfortunately for Ockham, there was an additional inferential step: abstract qualities are not mind-independent entities but written and spoken terms. These terms can have, at best, a fictional existence; that is, they are products of our minds, singular qualities of thoughts. But there is no need to go on a metaphysical spree to reify and multiply these qualities.

Ockham never denied the existence of universals, so the famous Razor was never used for denying but rather for suspending judgment about the entities. Nowadays, we can usually find the following formulation of Ockham’s Razor: “Don’t multiply entities beyond necessity.” Alas, Ockham never said this. In the words of the renowned philosopher Paul Vincent Spade:

“Although the sentiment is certainly Ockham’s, that particular formulation is nowhere to be found in his texts.” (2019, Stanford Encyclopedia of Philosophy)

Spade further points out that one particular sentence in Ockham’s Commentary on the Sentences of Peter Lombard can be the only suitable candidate for the “real” Razor, which he translates as:

“For nothing ought to be posited without a reason given, unless it is self-evident (…) or known by experience or proved by the authority of Sacred Scripture.” (2019, Stanford Encyclopedia of Philosophy)

In other words, Ockham stated three cases in which it makes sense to posit entities. In every other case, one is better off without speculative ontological categories. He did not label his principle as a “Razor” either.

Some three centuries later, Irish Franciscan friar and commentator of Ockham’s works, John Punch, introduced the phrase “Ockham’s Razor” to convey the idea of slicing through or cutting away bulky and unnecessary ontological categories and commitments. Thus, in due time, Ockham’s Razor became inseparable from the principle of ontological parsimony. This means that ontological simplicity is preferred when assessing competing hypotheses or theories. In this way, Ockham’s Razor favors simplicity as a theoretical virtue.

Can We Apply Ockham’s Razor in Scientific Inquiry?

Given that all scientific theories and hypotheses have some kind of ontological baggage, it seems quite natural to apply Ockham’s Razor when we are unsure of the direction of our scientific inquiry. In other words, it is intuitive to consider simpler hypotheses as more probable explanations of phenomena. Consider Galileo’s verdict on the rivalry between Ptolemaic and Copernican astronomy:

“Nature does not multiply things unnecessarily; that she makes use of the easiest and simplest means for producing her effects; that she does nothing in vain, and the like.” (1632/1962: 397)

Galileo, the godfather of the scientific method, in the Early Modern period wanted to tell us that Copernican astronomy uses fewer ontological commitments than Ptolemaic astronomy. Copernicus justified the heliocentric system because it was a more elegant and harmonious solution than Ptolemy’s geocentric system.

Why? Well, Ptolemy and his successor handled every counterevidence to their system by introducing more entities and commitments. For instance, Ptolemy introduced epicycles and deferents to explain the retrograde planetary movement, but his successors kept adding smaller epicycles to large ones for every failed prediction.

Throughout the Early Modern period and the Enlightenment, Ockham’s Razor became a symbol of the quest for a deeper, fundamental heuristic in experimental inquiry that would result in a clean-cut inventory of what there is. This is the period when excessive ontological baggage became tightly intertwined with speculative and non-rigorous science, thereby marking the first schism between metaphysics and science.

Along these lines, Lavoisier, the pioneer of modern chemistry, wrote the following about the hypothetical existence of phlogiston:

“If all of chemistry can be explained in a satisfactory manner without the help of phlogiston, that is enough to render it infinitely likely that the principle does not exist […]. It is, after all, a principle of logic not to multiply entities unnecessarily.” (1862: 623–624)

Phlogiston was conceived as a substance without color, smell, taste, or mass released into the air during combustion and other chemical reactions. The removal of this substance supposedly resulted in the true nature of things. For instance, the true form of metal was believed to be revealed once it became calx and lost phlogiston through the heating.

However, Lavoisier was among the first chemists to refute the phlogiston theory, which could not have been done without invoking the famous Razor. Lavoisier’s experiments showed that combustion involved the combination of a substance with oxygen from the air, not releasing a mysterious substance. This led to the development of modern chemical oxidation theory and the elimination of unnecessary ontological entities like phlogiston.

The Relevance of Ockham’s Razor in Contemporary Philosophy of Science

Despite its historical importance, Ockham’s Razor is not without its critics nowadays. Some argue that its preference for simplicity might lead to the oversight of complex yet crucial aspects of phenomena, thereby ridding us of scientific truth.

To put it mildly, we may be biased toward simpler theories and hypotheses due to our “limitations” as humans. These limitations deter us from finding the truth that must be out there (as Scully and Mulder tried to convince us in The X-Files).

Nevertheless, the plea for fewer ontological entities and commitments has some merit. The burgeoning subfield of philosophy of science—philosophy of artificial intelligence (AI)—may profit from emphasizing Ockham’s Razor.

Consider the daily problem that AI engineers face, namely overfitting models to data. This happens when models have been trained too well on their training data to the point where they capture the underlying patterns and the noise within the training dataset. When tested against novel data, models then perform poorly.

Ockham’s Razor tells us that we should prefer simpler models over complex ones to trace the relationship between the dataset and parameters within models. This improves the efficiency and interpretability of AI systems and aligns with the broader scientific quest for elegance and simplicity in understanding complex phenomena. Therefore, Ockham’s Razor can be relevant to the methodological issues we struggle with today.

References

Galileo, G. (1632/1962). Dialogue Concerning the Two Chief World Systems (trans. S. Drake). Berkeley.

Lavoisier, A. (1862). Réflexions sur le Phlogistique. In Oeuvres (Volume 2). Paris.

Spade, P. V. & Panaccio, C. (2019). William of Ockham. In E. N. Zalta (ed.), The Stanford Encyclopedia of Philosophy, https://plato.stanford.edu/archives/spr2019/entries/ockham/