//****************************************************************************//
//************* Whewell and Hypotheses - September 5th, 2019 ****************//
//**************************************************************************//
- Alright, another day, another quiz!
--------------------------------------------------------------------------------
- Okay, today's paper was by William Whewell (pronounced "Hue-wool")
- Whewell (1794 - 1860) was an English polymath and professor at Cambridge who did work in a BUNCH of fields - physics, math, Anglican theology, geology, history, and (as we'll see today) philosophy of science
- He actually coined the term "scientist," along with several others we still use today (like "linguistics" and "electrode")
- He did all of this at a time when specialization was becoming the norm, bucking the trend to become one of the last great polymaths
- In this paper, Whewell claims that previous philosophers had fundamentally misunderstood how induction worked - how?
- In a word, hypothesizing! In section 2, he points out that the "connecting rule" when we do induction isn't often obvious - instead, we need to creatively come up with a new rule on our own that'll connect all the facts together!
- Whewell says that these hypotheses often involve going BEYOND the evidence - we can't "see" gravity, so when Newton came up with his theory of gravitation he had to make it up and create something new. As Whewell says, it's the "string connecting the pearls"
- In section 3, for instance, Whewell pulls a syllogism from Aristotle:
1. All planets move in elliptical orbits
2. Mercury, Venus, and Mars are planets
3. Therefore, Mercury, Venus, and Mars move in elliptical orbits
- Here, he notes that proposition 1 doesn't just pop up out of thin air, and it doesn't become obvious from looking at the data, whereas many people take coming up with this for granted
- Aristotle might've worked hard trying to prove planets move in elliptical orbits, but Whewell points out the hard step here is imagining the very idea of "elliptical orbits" at all!
- The ancients originally imagined all the orbits were circles; when it became clear that planets didn't move like that,they instead came up with "epicycles" where planets would move in combinations of circular orbits - circles around points on a bigger circle, and so on - and it actually did kinda work!
- So, for Kepler to first imagine planets could have non-circular orbits at all was a HUGE step!
- Once these "inventive steps" are made, Whewell points out that it's often taken for granted, and once we have an idea of how something works in our heads it's hard to change it, to the point where "epicycles" now seem ridiculously out-of-date to us "moderns"
- At the same time, he does say that these theories have value as building blocks, and we shouldn't mock them or think they were totally without merit - after all, they must've had some explanatory power if they were used
- In Whewell's mind, science isn't just a meticulous task, but requires bold, creative invention to come up with new guesses and hypotheses
- Obviously, these hypotheses and "conjectures" (section 6) should be based on the data and try to explain the data, but it does involve going beyond the data itself
- Once we have these hypotheses, we need to test them (per section 7): we need to compare what our hypotheses predict with the facts, and be "ready to abandon [our] invention as soon as it appears that it does not agree" with the truth
- Here, these tests are focused on disproving the hypothesis, and beating the heck out of it to make it hold up
- What makes a hypothesis TRUE, then, is if after this beating we can also use the hypothesis to PREDICT new things, and ESPECIALLY if our hypothesis generalizes and can prove unexpected things
- He calls this highest form of proof the CONSILIENCE OF INDUCTIONS, where the hypothesis can make accurate predictions for things we weren't even thinking about when we made the theory, or didn't even think were related (e.g. Newton's gravitational theory and tides)
- This looks especially good if the theory makes predicting these things SIMPLER than current techniques
- Conversely, Whewell thinks we should be skeptical about hypotheses where we have to make defensive hypotheses just to defend old predictions as new data comes to light - if we have to keep adding to the hypothesis as more data comes to light, he think that's a sign the hypothesis is being strained
- Notice that the title of this paper was "Hypothetico-Deductivism" - Whewell's method involved induction AND deduction
- We use induction to come up with a hypothesis, and then deduce "okay, if this is true, we should expect to see THIS" - and then we compare that with reality!
- So, the deductive step here is trying to figure out what the full implications of our hypothetical claims are, and then making sure that matches up with reality
- Notice here, too, that for Whewell a true theory is one that accurately represents the way the world actually is, which he proposed doing by seeing if the theory makes accurate predictions
- The epicycle theory of orbits was false, but it made pretty accurate predictions - enough for sailors to navigate to specific cities from the stars
- How can we make an atom bomb if our theory isn't true, right? I mean, that'd have to be a massively huge coincidence, so this proves our atomic theories are ironclad, right?...right?
- ...we'll come back to this question several times throughout the course
- So, Whewell's account of science here is VERY close to the scientific method they teach in schools nowadays and that you've probably read in ten billion different physics textbooks
- "...we're going to complicate this story a LOT - it seems crazy to think there's only 1 scientific method, after all - but it is generally a pretty modern form of what we're still doing today"
- Alright; next week we'll do our last "historical" reading from John Stuart Mill, and then we'll get into the modern day - see you then!