Rhetorica:
Of Bacon;
or, The Scientific Method
By Gabriel Blanchard
Remember: Not all science is equally rigorously conducted; if you are testing a hypothesis, you are doing science—it might just not be very good science.
A Tale of Two Bacons
Much of what we call “the scientific method” is owed to an English philosopher of the Bacon family who lived several centuries ago. A devoutly religious member of a family noted for its loyalty to the English crown, and sympathetic to the cause of ecclesiastical reform, he was educated by an Oxonian1 and had a great liking for Aristotle. He wrote a number of scientific treatises, laying emphasis on the importance of acquiring knowledge through experimental observation, and he had a major influence upon the subsequent shape of education.
That seems to be all that one can truthfully say about both men. Whether Friar Roger (c. 1219-c. 1272) or Sir Francis (1561-1626) was the Bacon in question there is a choice which may be left up to the reader’s preference.2, 3 In any event, what both contributed to, and what we have to discuss today, is one of the principal forms of inductive reasoning; namely,
The Scientific Method
At its most basic, the scientific method means simply “testing your guesses about the quantifiable world.” If you are testing a guess—something we all do every day—you are, in one sense, doing some science. For most of us, most of the time, the differences between ourselves and, say, a cancer researcher at Johns Hopkins are that (a) less is riding on whether we guess correctly, and (b) the expertise and equipment we bring to both guessing and testing don’t cost $267,000. Apologies to any readers who do conduct their everyday lives with such a ferocious commitment to getting things exactly right (although the present author feels at the same time bound to submit that this ferocity is not sustainable in the long term, either psychologically or financially).
Contrary to popular belief, there is no single formulation of the scientific method generally recognized as the authoritative one. You’ll find minor differences in the naming, numbering, and (rarely) even the ordering of the steps it consists in. However, the basic concept of “testing a guess” underlies all versions, and while there’s room for variation and adaption in what “testing a guess” looks like, it’s a coherent thing; the variations and adaptions aren’t done merely because the experimenter felt like it. It is therefore tolerably safe to summarize the procedure; generalizing, we may say that the following six steps are standard:
I. Observation
II. Hypothesis
III. Prediction
IV. Experiment
V. Analysis
VI. Conclusion
Let’s unpack each of those a bit. What do they each mean? Why are these (or something as like them as makes no difference) the backbone of Baconesque induction? Are they really all indispensable?
I. Observation.
This means discerning quantifiable facts, and identifying something unexplained about them—some question you wish to answer, or to help someone else answer.
Discerning quantifiable facts is crucial, not because there is no other kind, but because other kinds are not really susceptible to this type of analysis. (You strictly can discover whether someone will respond in kind if you tell them “I love you”; however, to do so in a spirit of mere investigation rather than of sincere revelation is frowned upon, and, once discovered by the relevant parties, has a tendency to change their answer the next time the question is posed.) Quantifiable facts are those to which we can assign quantities: breadth, weight, density, wavelength, boiling point, number of dorsal fins, etc. This, of course, is why mathematics is the key to all the sciences.
As for identifying a question you want to answer, this is not an obligation sensu stricto; you’re free to notice facts and just sigh contentedly about them. However, if you don’t identify a question, … what are you doing here? People-watching isn’t the worst hobby in the world, but it isn’t widely appreciated inside laboratories.
II. Hypothesis.
This means coming up with a possible explanation of the unexplained facts, a possible answer to the question. For obvious reasons, a hypothesis therefore needs to (a) explain the mysterious facts, and also (b) not be inconsistent with anything we already know.
An exemplar will do us good. Here we have Thag, a scientist from approximately seventy thousand years ago. He has observed that he sometimes thoughtlessly lets go of his mammoth steak, and on those occasions when he does, it often falls to the ground. In fact, he can’t think of a time it’s ever just stayed up at eating level. And come to think of it, the same thing is true of his spear. Thag takes thought, and hypothesizes thus: There is a malevolent force, which I will name after my bullying older brother, Zok. This Zok force follows me around, and whenever I let go of something, this force zoks it to the ground. This is, so far, an admirable hypothesis: It contradicts nothing that Thag knows, and it does explain why his mammoth steaks and his spear keep hitting the dirt.
III. Prediction.
This means thinking out what else would have to be true that we have not yet observed, if the hypothesis from Step II were correct. Ideally—this isn’t always possible—hypotheses should have a few different, very specific predictions, including some variations according to circumstance; that kind of variety makes Step IV easier. But this is just a perk—a single prediction is enough to be following the scientific method.
Thag, as it happens, has a single prediction with many applications. His prediction is: If the Zok force is real, every other thing I let go of in midair will also be zokked to the ground.
ὁδὸς ἄνω κάτω μία καὶ ὡυτή.
The way up and the way down are one and the same.
IV. Experiment.
This means first devising, and then implementing, a method to detect whether the predictions we have made are true.
Unfortunately, depending on the complexity of our hypothesis, sometimes we devise an experiment but cannot carry it out.4 Even this has value, since, if we share the design of the experiment with other people, one day others may be able to perform it instead; however, unless we can devise an experiment we are capable of performing ourselves, our own scientific journey halts here.
Some of our readers may feel that Thag has not cut a very impressive figure thus far, but the experiment he devises is undeniably cunning. He collects five objects—a gazelle steak, a rock, a longer and thinner rock, a large beetle, and a dead leaf—and drops each one of them in turn.
V. Analysis.
This means measuring the facts that result from conducting the experiment to see whether they support the hypothesis or tell against it—or, more puzzlingly, do neither. The catch about this part of the scientific method is that, an awful lot of the time, we’re hoping for one answer rather than another, which can motivate our interpretations, something we must be on ruthless guard against.
Luckily for Thag, his experiment’s results are clear and simple enough that he could not seriously interpret them according to his preferences; luckier still, he has been wondering out of genuine curiosity, with no preferences about the answer. The gazelle steak, the rock, and the rock all conform with the Zok Hypothesis; the beetle, on the other hand, flies off, which doesn’t match at all. Moreover, the result from the dead leaf is ambiguous: it does eventually reach the ground, but it flutters and drifts about for so long on its way there, Thag ultimately decides that it too does not operate as he would have expected on hypothesis.
VI. Conclusion.
This means first synthesizing and summarizing the information from Steps IV and V, and then sharing the summary with others, especially other researchers in the same field.
Usually, the summary will take one of three forms: A. “I was right”; B. “I was wrong”; or C. “I’m confused.” (There can also be combinations, in-between answers, and so on, especially if the hypothesis is a complicated one.) Any summary, when shared, constitutes an invitation to fellow scientists to perform the same experiment and share their conclusions. Sometimes the initial theorist will get a B conclusion, but the input of others makes it clear that that B was an exception, not the rule. Or the same thing could be shown about an A conclusion; you don’t know until you try. And no matter what the conclusions are, the more of them you have, the better, because that aggregation will do one of two things:
- produce a most-common-on-average answer, or
- not,
and again, either of those results helps build up your knowledge of the world. If the “average” answer is the A kind, then you can have more and more confidence in your initial hypothesis being correct as the A’s come in. If you get any other answer, that’s also helpful—it tells you that your hypothesis is incorrect or, at least, flawed. That is an advance in your knowledge; but it may not feel like one. Let’s linger briefly on that to close.
Proud Me No Prouds
The Zok Hypothesis is obviously a crude, anthropomorphic theory of gravity; but it is recognizably a theory of gravity. It doesn’t take the movements of living things into account that would explain the beetle, and it lacks the fluid dynamics component that would explain the leaf, but it’s not incorrect that things fall down. The Zok Hypothesis is about, hmm, 25-30% right, and Thag has made a tiny but real advance by coming up with it. Now that he’s gotten some B results from his experiment, he can go back to Step II and tinker with his hypothesis till it supports new predictions, and then he can test those—and so on.
Unluckily, we are frequently very silly creatures, and discovering that we have been wrong at all—even under circumstances where no one could reasonably hope to be right—can be an exceedingly unpleasant jolt. The present author regrets deeply to inform you that, to his knowledge, there is nothing to do with this electrocution of the ego but wait for it to stop.
A few meditations may (or may not) help it to pass more and more quickly: the fact that nobody gets everything right all the time; the fact that this is an advance in your knowledge, even if it’s not an advance in your pride; the fact that you have probably unfairly made fun of someone else’s ignorance before, and by getting some “karma” for that now, you’re hopefully correcting yourself spiritually a little bit. This also helps highlight some of those general lessons we discussed at the beginning of the Rhetorica series—it may be easier to see why magnanimity is important in this moment when you could use some, and it’s certainly a chance to grow into the “little trinity” of intellectual virtues, since it takes courage to maintain honesty and clarity while your brain still stings. They’ll keep coming back.
1This is the less-than-transparent adjectival form of “Oxford”; something or someone from, or associated with, Oxford is “Oxonian.” (Curiously, this adjective does not seem to come from the Middle English form of the city’s name, which was Oxenford or Oxnaford; rather, it appears to derive from the Latin name for the place, Oxōnia—though this likely does come from Oxnaford.)
2There is a better case for its being Francis than Roger Bacon. The Franciscan friar’s enthusiasm for experiment, though not uncommon among Medieval scientists (cf. Alhazen, al-Biruni, St. Albert the Great, and Robert Grosseteste), was less thoroughly applied than that of the Elizabethan jurist; and, while Sir Francis Bacon has influenced the procedure of all the sciences (and thus education in general), Br. Roger Bacon’s impact only went as far as persuading the Medieval university to add optics to its core curriculum—though that’s nothing to shake a stick at.
3Although they were not close relatives—one lived in the thirteenth century and one in the later sixteenth and early seventeenth—it seems that Roger and Francis did not have the same surname just by coincidence, but were scions of the same lineage, many generations apart. The Bacon family were armigerous Normans who settled in northwestern France before immigrating to southwestern England after the Conquest; the arms in question were Gules six roses Argent seeded Or, i.e., a red field with six white roses (arranged in a downward-pointing triangle), each rose having a golden center.
4There is an amusing and instructive example of this in the history of astronomy. The first astronomer of note to suggest a heliocentric model of the universe was neither Galileo nor Copernicus, but a Greek scholar named Aristarchus who lived in the third century BC. His hypothesis failed to gain favor at the time because one of the predictions built into it was stellar parallax, something no one observed with the equipment they possessed at the time. Aristarchus’ counter-argument was that the stars were much further away than anybody else thought, so that the actual detectability of parallax sank to zero; very understandably, the rest of the astronomers of his day more or less cocked an eyebrow and said “How convenient,” and went right on teaching the geocentric model that accorded better with the facts at their disposal. This is partly why even an accepted scientific theory, though it might be folly—even, under some circumstances, reprehensibly negligent—to reject it, remains a theory all the same.
Gabriel Blanchard is CLT’s editor at large. He lives in Baltimore, MD.
If you enjoyed this piece, check out our series on the Great Conversation, the intellectual exchange between scientists, statesmen, philosophers, artists, theologians, mathematicians, and historians that’s been going on since the Bronze Age. Thank you for reading the Journal.
Published on 15th May, 2025. Page image of a “full Irish” breakfast, conceptually akin to the “full English” but differing in details; photo by flickr user jules:stonesoup, used under a CC BY 2.0 license (source).
