Cross-post: The idea of a University

The Idea of a University in Nine Discourses
by John Henry Newman (1858)
available online

At a time when the proper mission of a university has been obscured by commercial and ideological interests, we can with profit consult the classic lectures on this topic delivered by Cardinal Newman to commemorate the establishment of a Catholic university in Dublin.

It is unfortunate, as Newman points out, that English lacks a convenient word for what he means as the distinctive excellence of the intellect, the equivalent of what “health” is for the body, because this is what a university education is meant to cultivate.  Intellectual cultivation might aid professional success and moral refinement, but it is a separate good worthy of pursuit in itself.  Newman refers most often to two particular facets of the properly formed mind.  First there is what one might call a philosophical enlargement, an appreciation for the validity and proper limits of each discipline.  Second, there is what he sometimes calls discipline of the mind, the habit of precision and systemization.

Book review: The Metaphysics Within Physics

The Metaphysics Within Physics
by Tim Maudlin (2007)

In this collection of essays, philosopher Tim Maudlin makes an extremely strong case against the Humean philosophy of physics.  This position, credited to Hume and popular among contemporary philosophers, posits that the universe is completely described by physical facts about localized quantities at each different place and time.  As Maudlin points out, this doesn’t match current scientific theory or practice.  The state of a system with spatially separated but entangled parts cannot be factored into purely local pieces, and science requires not only facts but also scientific laws.  Laws are not reducible to facts; if they were, they couldn’t do their work of making predictions and answering questions about counterfactual scenarios.  I have in the past distinguished the “Platonic” and “Aristotelian” ways of thinking about laws of nature:  the former speaking of them as having some independent existence, the latter regarding them as being embedded in the natures of existing things.  Maudlin is apparently a “Platonist” in that he takes laws to be completely primitive, but nearly everything he says would also be endorsed by an “Aristotelian” with the suitable reinterpretation of the idea of physical laws.

I take the polemic against the followers of Hume to be the major point of the book, but Maudlin makes several other interesting observations along the way.

He suggests that we are able to assign causality intuitively because the laws of nature approximate a particular “quasi-Newtonian” form in which separable subsystems each have their own “inertial” operation that will happen as long as nothing interferes, plus an equivalent of force laws to describe how objects can influence each other.

He argues that we have no reason to believe in “simple” properties or relations, meaning properties/relations that refer to nothing but their subjects/relata, and so we shouldn’t build our metaphysics out of these.  He proposes gauge theories as a counterexample, in which vectors on fibers at different base points can only be compared via a connection and a connecting path on the base manifold.  As a proof, this would be begging the question, because the spaces that make up the fibers in a fiber bundle usually have an inherent homogeneity to them which “simple” qualities (if any exist) would break.  However, the proposal is valuable for shifting the burden of proof.  Do we have any reason to believe in simple properties and relations?

Maudlin believes that time actually “passes”.  He’s not a presentist, so he’s not running afoul of relativity in this.  I usually say I don’t believe time passes for the same reasons Maudlin doesn’t believe that time “flows”:  what could it flow with respect to?  Doesn’t flow just mean change with respect to time?  But if “passing” isn’t the same as “flowing”, what does it involve?  Maudlin finds it difficult to explain.  It includes an asymmetry between past and future.  This is supposedly not the whole of what it means for time to pass, but Maudlin’s arguments are mostly aimed toward showing that such an asymmetry is plausible (and not reducible to differences in entropy).  I don’t have any objection to such an asymmetry, so it’s not clear how much Maudlin, who asserts that time passes, and I, who assert that it doesn’t, really disagree.

Book review: The Individual and the Cosmos in Renaissance Philosophy

The Individual and the Cosmos in Renaissance Philosophy
by Ernst Cassirer (1963)

Historians of Renaissance and early modern philosophy often try to give a unity to their subjects by framing the creative elements of these periods as engaging in a revolt against “scholasticism”.  However, this only gives as much intelligibility to the Renaissance as is granted to its foil, and historians usually assign scholasticism any negative quality needed to keep the narrative going;  it can be mindlessly dogmatic or aridly intellectual or both at once, despising all nature or assigning fanciful hierarchies within it, servile or unfaithful to Aristotle, holding an opinion of man that is irrationally low (when the opponent is humanism) or high (when the opponent is science).  Cassirer tries to fit his study into this standard narrative, but he provides a great deal of interesting material, so that a more interesting story begins to emerge.

Cassirer’s exemplary Renaissance philosopher is Nicholas of Cusa, the idiosyncratic Christian neo-Platonist who smashed the medievals’ hierarchical universe to stress the incomparability of God, the Absolute and Infinite, the confluence of opposites.  He imagined the Earth in motion (which he seems to suggest is relative) in an infinite universe with no center but God, a physical infinity to which corresponds the intentional infinity of the human mind–whose operation is now conceived primarily in terms of measuring and comparing.  From Cusanus, Cassirer expands to cover a number of other characters:  Platonists like Marsilio Ficino and his Florentine Academy, humanists like Petrarch and Pico della Mirandola, proto-Hegelians like Charles de Bovillus, and those climbing toward a scientific approach to nature like Leonardo and Galileo.

Several things become clear.  The recovery of Plato’s dialogs made Plato a rallying point against Aristotle for a rather diverse group of thinkers.  Why should this be?  Petrarch’s preference for Plato over Aristotle and his scholastic followers was primarily aesthetic and therefor frivolous.  Most of the others had disagreements with Aristotle but ones that hardly seem to take them outside of the orbit of scholasticism, i.e. not farther from Aquinas or Ockham than these two are from each other.  One often encounters an assumption that separation from scholasticism means approach to secularism, an assumption popular because it is so congenial to both secularists and scholastics.  I’ll  therefore mention that most of these thinkers gave every impression of being ardent Christians.  And yet, they did consider themselves at war with the Aristotelian schoolmen.

Then Cassirer, in the final chapter, gives a revealing fact.

To understand the transformation that takes place with the beginning of the philosophy of the Renaissance, we must keep in mind this opposition, this tension, which already existed in the medieval system of life and learning.  Despite all the attacks it had suffered in the classical systems of Scholasticism, the theoretical foundation of Averroism seemed to be completely unshaken in the 14th and 15th centuries.  For a long time, it was the reigning doctrine in the Italian universities.  In the actual academic citadel of Scholastic studies, in Padua, Averroistic doctrine maintained itself into the sixteenth and seventeenth centuries.  But gradually, a counter movement emerges ever more clearly.  Characteristically, this counter movement is by no means restricted to the environs of the school, but rather receives its strongest impulses from other quarters.  The men of the new humanistic ideal of culture and of personality are the first to sound the call to do battle against Averroism.  Here, too, Petrarch leads the way…The artist and virtuoso who rediscovered the inexhaustible wealth and value of ‘individuality’ now sets up his defenses against a philosophy that considers individuality to be something merely casual, something purely ‘accidental’.  And Augustine becomes his guide in this battle.

Well, if Averroism dominated the Italian universities (a fact which is new to me), and that’s what the humanists meant by “Aristotelianism”, then it becomes very clear how the writings of Plato–with their support for personal immaterial immortality–could serve as a philosophical rallying point to the opposition, and also how a literary movement devoted to individuality would be so adamantly a part of this opposition.  Replace “the Renaissance was a rebellion against scholasticism/medievalism” with “the Renaissance was a rebellion against Averroism”.  It would take a good deal of confirming evidence before we believe it (and Cassirer continues with citations to attacks on Averroism from many of the Renaissance greats from Cusanus on), but at least this new narrative makes sense.

The last chapter (which is by far the best of the book) also relates the Renaissance’s stumbling toward the scientific method.  The misfires are particularly informative.  One finds that prizing experience over a priori reasoning isn’t enough, at least given a medieval credulity to reports and a tendency to express observations in magical categories.  A commitment to a believe in a universal rational order of the universe isn’t enough; that led to painstakingly systemized astrology.  (Astrology made a big comeback in the High Middle Ages / Renaissance with the influx of pagan and Muslim learning.  Once again, the Church and the humanists were on the same side in the fight against it.)  Cassirer thinks what was missing was a mathematical approach to nature, and this came from scientist-artist like Leonardo da Vinci and their attention to form.  However, if focus on form is what you want, then Aristotle is your man.  And yet, the great men of the scientific revolution like Copernicus, Galileo, and Kepler were Platonists.  What gives?

The main difference between Plato and Aristotle is that the former makes the intelligibility of the world transcendent, the latter immanent.  Imagine you were Galileo trying to understand how bodies fall.  For an Aristotelian, action follows being.  To learn how a body falls, you must first ask what is the nature of that body, given by its substantial form.  You would not expect there to be a general rule about how things fall, because different things have different natures and hence different principles of motion.  (cf. Nancy Cartwright’s “dappled world”)  The idea of general laws of motion is much more natural in a Platonic/transcendent framework.

So there’s a story that makes quite a bit more sense than the standard story.  The fight against Averroism promoted Platonism, and Platonism gave us science.  That’s the intellectual story of the Renaissance.

Book review: The Order of Time

The Order of Time
by Carlo Rovelli (2018)
also his
Time in quantum gravity: an hypothesis, Phys Rev D 43, 442 (1991)
Statistical mechanics of gravity and the thermodynamical origin of time, Class Quantum Grav. 10 1549 (1993)
Relational Quantum Mechanics, Int. J. of Theor. Phys. 35 1637 (1996)

I saw Carlo Rovelli, inventor of loop quantum gravity, give a talk once.  I believe it was at GR22 in Warsaw.  It was my first exposure to his general philosophy of doing physics.  Rovelli thinks that questioning the core insights of quantum mechanics and general relativity is by this point an unpromising strategy for theoretical physics.  Our task is to extend and synthesize them.  Like his fellow Italian Thomas Aquinas, Rovelli is a synthesizer; by my count, in this book he synthesizes Anaximander, Aristotle, St. Augustine, Newton, Leibniz, Boltzmann, and Einstein.  All around his surprising claim that time, at the most fundamental physical level, does not exist.  Surprisingly for a work of science popularization, he ends up agreeing with phenomenologist philosophers who claim that the “lower-case t of physics” doesn’t capture the human reality of time.  Rovelli agrees that the essence of time is to be found in human subjectivity, and claims that physics itself leads to this conclusion by murdering the “lower-case t of physics”.

The surprising powerlessness of scientists in a culture war

(Expanded from a comment on the most recent post)

The modern world is supposedly built around a scientific view of the world.  If so, that would make scientists our official prophets.  (Some say “priests”, but the role of the scientist is more analogous to that of the prophet than that of the priest.)  One would think that, to capture the culture, having scientists on one’s side would be among the most valuable possible assets.  But that’s really not the case.

Don’t get me wrong.  Having scientists from one’s group is a good thing because science is worth doing:  the truths it reveals are worth knowing, and the discipline it teaches is worth having.  The attempt to use science in a culture war only corrupts it and produces pseudoscience.  Just as a man cannot decide to learn Stoic detachment for the purpose of financial gain, scientific truth is one of those goods that can only be pursued successfully if done so for its own sake. As Bertrand Russell said about philosophy, science will answer only its own distinctive questions.  However, scientists can do very little on their own to help dominate a culture.  Having most scientists on one’s side is an effect of winning a culture war rather than a cause.

Consider

1) One certainly hopes that scientific discoveries do not depend on the prior beliefs of researchers. Therefore, stuffing sympathetic personnel into a field shouldn’t affect its conclusions.  (If it does, it’s not real science.)  A person with different loyalties might indeed investigate different questions.  Religious/political demographics probably do affect research programs in the social sciences, but I doubt they are of much relevance to the real sciences.  (Yes, I’m letting my prejudices show.)

2) Nor do scientists get a privileged role in interpreting their own discoveries. Nobody cared that Kepler took his model of the solar system to be itself a model of the Trinity.
Fermat, Leibniz, Maupertuis, and Euler all thought the principle of least action is a sign of God’s perfection. Eighteenth century French atheists claimed to base their worldview on Newtonian physics but took no interest in Newton’s own wacky Arian millenarianism. Descartes thought his physics had demolished 17th century materialism (his mechanical philosophy devised to emphasize how distinct are mental phenomena), just as Heisenberg thought his physics had demolished 19th century materialism (by overthrowing its epistemology), and Lemaitre thought he had destroyed the materialists’ eternal universe.  Maxwell used the indistinguishability of elementary particles (atoms, for him) to advance a novel design argument.

Today, the fact that many scientists thought their discoveries were irrelevant to–or perhaps even supportive of–Christianity is regarded as a historical curiosity. The narrative imposed on the history of science since Copernicus is of the great liberation from Christian superstition. This narrative comes largely from French men of letters rather than scientists themselves; the latter having been converted to it not earlier than the late nineteenth century.

By the way, this is one reason I find seventeenth century natural philosophers so fascinating.  It’s not that their beliefs about the meaning of their work were necessarily truer than the later Enlightenment and contemporary views, but that they could be so different, showing how much one’s metaphysical and historical presuppositions color how one does something so apparently nonpartisan as interpret scientific theories.

3) In any case, the philosophical interpretation of scientific theories is I think much more difficult than most people realize. Those who think it’s easy to read ontology out of physics or biology are most often reading their presuppositions into it. After nearly a century, many physicists are not shy in saying that we still don’t really understand quantum mechanics, even though it’s straightforward to use, most likely because some unacknowledged metaphysical prejudice is still being worked out of our system. As another example, that parts are ontologically prior to their wholes is an assumption which detailed scientific study of cells, atoms, etc can neither confirm nor disprove.  Plato and Aristotle believed wholes to be ontologically prior, while I find the whole idea of ontological priority suspect.

4) Non-westerners encountered Western science, mores, and overwhelming technological supremacy all at once, and it was natural that they would sometimes regard them as a single package, but for Westerners it is different.  Roughly our history is as follows.  In the seventeenth century, Christians of various stripes carried out the scientific revolution.  In the eighteenth century, atheists and deists used the success of science as an argument against Christianity.  In the nineteenth century, science continued to advance, and with the Industrial Revolution, the new knowledge was now changing people’s material lives in obvious ways.  Among the ranks of scientists and inventors, there was still a large diversity–Christians to atheists and everything in between.  Only in the twentieth century did science clearly come to be dominated by atheists and Jews, long after some other fields had so aligned.  So the West has seen science change hands and is less liable to see it as the unique genius of some faction.  Being an atheist doesn’t automatically make one more “scientific” than Pascal or Maxwell.

5) Persecuting scientists doesn’t hurt one’s reputation unless one is already weak.  The Left paid no price for the murder of Lavoisier or for interfering with genetics research in 20th century Russia and 21st century America.  The weapon of getting to tar people as “anti-science” is not one that scientists themselves control.  To be clear, I’m not recommending anyone persecute scientists, just pointing out a sad fact that one can get away with it if one’s social standing is strong.  The example of Soviet science shows that one can even remain world-class in some fields (Soviet mathematics was top-rate, and of course they got most of the “firsts” in the space race) while descending into crackpottery in others.

book review: Being as Communion

Being as Communion: a Metaphysics of Information
by William Dembski (2014)

I became aware of this book via a comment by Kristor at The Orthosphere, and it sounded interesting.  Dembski is one of those intelligent design eccentrics, and being one of those with strong materialist prejudices against which the book is written, I probably would not have come to it on my own.  I’m glad I did read it, because I learned that intelligent design doesn’t claim what I thought it did.  Its claim is that life arose from exterior design or teleological features of matter unknown to physics.  So the elan vital is one possibility they’re raising.  Not that I’m more sympathetic to Bergsonian weirdness than I am to creationism, but once again I find that hostile sources are never, ever reliable on their opponents’ beliefs.

By the way, Dembski describes an interesting paper by John Hawthorne and Daniel Nolan on what teleological causation would be like.  The key is that it would be nonlocal in configuration space, giving higher than locally expected probabilities to system changes that are on the path to some attractive goal.  This nonlocal dynamics is to be contrasted with the nonlocal kinematics (the state spreading over/sampling a range of possible property values) of existing quantum mechanics, although the phenomenon is still reminiscent to me of quantum tunneling.

In this review, I will concentrate on what I take to be Dembski’s two core claims, namely his advocacy of an ontology in which information is primary and his arguments that his mathematical work on search algorithms poses a problem for natural selection.

book review: The Philosophy of Space and Time

The Philosophy of Space and Time
by Hans Reichenbach (German: 1927, English: 1958)

pseudo-problems arise if we look for truth where definitions are needed.

–pp. 15

This is the classic work on its subject, but I have only now read it, having fortuitously come across a used copy in Bruised Books in Pullman.  I come to this book late, deterred by the prejudice in my field that books on general relativity written before the 70s are boring and obsessed with transformation rules, but I wish I hadn’t waited so long, because it is a masterpiece.

Book review: What makes time special?

What makes time special?
by Craig Callender (2017)

This book was a joy to read.  Its question can be formulated as follows.  All peoples share a conceptualization of time (as distinguished from our direct sensory experience) that Callender calls “manifest time”–of time “flowing” with a special time “now”, etc.  None of these features appear in physics–Aristotelian, Newtonian, relativistic, or quantum–so where does our sense of manifest time come from?  This question is broken down into parts.

1. Since “now” would be a special spacelike hypersurface of spacetime, it is established that the laws of physics do not justify singling out any particular foliation of spacetime.  This might seem obvious, but Callender wants to show that proposals to inject features of manifest time into physics fail in their goal.  One example is causal set theory.  Callender argues against the theory’s own founder–and I found the argument entirely convincing–that the flowing now in causal set theory is a gauge freedom with no physical significance.  Callender actually gives slightly  more credit that I do to arguments from quantum nonlocality.  He does show that there is only an issue here if one adopts the Copenhagen interpretation and insist within it that there must be some some fact of the matter about which of two spacelike observesrs collapses a wavefunction.  I would not be inclined to grant either premiss.  Even if one does and makes of it an argument for Lorentz’s interpretation of relativity as opposed to Einstein’s, Callender makes an important point.  Lorentz’s preferred frame is completely unobservable and so can’t have anything to do with our experience of time.
2. Physics does indeed give a reason to favor foliations of spacetime into spacelike hypersurfaces, a reason of “narrative”:  the world is more intelligible when cut up this way.  From other hypersurfaces (e.g. a plane extending in t, y, and z representing a “moment” of x), one could constrain the rest of spacetime, but only spacelike hypersurfaces provide initial data for a well-posed Cauchy problem:  the data on the initial slice is unconstrained, and the subsequent (and past) evolution is a continuous function of the initial data.  Ultimately, this comes down to the spacetime metric, which I would say expresses the causal structure of spacetime.  Callender mostly agrees, but his empiricism sometimes leads him to formulate things in (what seems to me) odd ways–that what’s real is observable events, which data can be compressed by fundamental laws of physics which happen to be hyperbolic equations with the same characteristic speeds, which makes it handy to speak of a causal structure of spacetime.
3. The explanation of manifest time must therefore be sought in human psychology.  There is a brief discursion into the argument over indexicals which all philosophy of time works address:  that our statements no more prove that there is metaphysically privileged “now” than that there is a privileged “here” or “me”.  I am skeptical of attempts to tease out metaphysics from grammar, but since it keeps coming up, I guess Callender had to address it.  More substantially, Callender argues against the intuitive belief that we have a direct experience of the present.  We are used to similar arguments from empiricists, e.g. “I have no particular experience of myself; therefore selves don’t exist.”  This argument isn’t quite that bad.  Callender has already given independent arguments that the now of manifest time is an illusion, so all he needs to do here is neutralize the counter-argument that we directly experience this putatively non-existent thing.  One might wonder (although the book doesn’t do so explicitly) if “now” is like the self in that, while not directly experienced, it must be posited to explain a perceived unity of experiences.  In fact, what is needed for this is just an experience of simultaneity, which seems to me rather unthreatening from a physicist’s point of view.  Nevertheless, Callender recounts experiments showing a degree of haziness in perceived simultaneity.  Some of it is interesting–I learned that we process sound faster than sight and that it’s possible to recognize that two stimuli are slightly non-simultaneous without being able to say which came first–but I don’t think it’s necessary for the main argument, which doesn’t hinge on how sharp our sense of simultaneity is.  There follows discussions of IGUS and evolutionary pressures to explain our experience of motion and greater concern over future vs. past pains.  In some ways, this is the most important part of the book, because it is here that manifest time is supposed to actually appear, but given my lack of interest in psychology I didn’t give it as much attention as the rest of the book.

Callender concludes with a criticism of what he sees as the philosophy of time’s disengagement from science and retreat into metaphysics.  In some ways, this complements my criticism of neo-Thomism’s retreat into metaphysics, but Callender drives home a point specific to the philosophy of time.  The point of this branch of philosophy is to explain our experience of time, but it is hard to see how metaphysical constructs that leave no trace on physics could be much use to phenomenology.  For example, even if there is an unobservable preferred frame, how could this have anything to do with how we experience time?

Book review: The Mind of God

The Mind of God:  The Scientific Basis for a Rational World

by Paul Davies (1992)

Davies articulates well several positions which I have also long held

• It is the universe’s combination of orderliness and contingency that metaphysics must explain.
• Multiverses and many-world versions of quantum mechanics won’t resolve the contingency issue, because the laws of physics themselves remain clearly contingent.  (To really get around contingency, you need to embrace full modal realism of the sort defended by philosopher David Lewis.  Surprisingly, some scientists, such as Frank Tipler, are willing to do this, making bizarre claims such as that the universe is a simulation and that mathematical consistency gives an ontological argument for everything.  Neither Davies nor I find this line of thought very credible.)
• The idea that the laws of physics can explain how the universe appeared from “nothing” assumes that these laws have some existence that transcends the natural world they describe, a truly perverse assumption.
• The principle of sufficient reason needs to be qualified to accommodate a contingent but orderly universe.  Some things but not others require explanation.

Davies’ own speculations force him most of the way through the classical cosmological argument for a necessary transcendent God, but he cannot fully accept classical theism because he sees strong, probably insurmountable, difficulties explaining how a simple, necessary God could interact with a contingent, complex universe.  (He is fair enough to point out that many other metaphysical positions, including versions of belief in transcendent laws of physics, have analogous interaction problems.)  He tentatively proposes a model, inspired by process theology, in which God establishes necessary probabalistic laws and potentialities but creatures/events decide by chance/”freedom” what is actualized.  I don’t think this will work, since the laws themselves are non-necessary.

There is an interesting attempt to apply ideas from information theory to give some precision to statements that the universe is “complex” but its laws are “simple”.  Since I don’t know anything about information theory, I enjoyed these parts of the book.

The Mind of God is intended for readers with no physics or philosophy background, but missing information is filled in succinctly enough to avoid boring other readers.

Book review: The Dappled World

The Dappled World:  A Study of the Boundaries of Science
by Nancy Cartwright (1999)

Cartwright proposes that we are led astray by our expectation that the world is united under a small number of laws.  She thinks it more likely that the world is a patchwork of distinct natures, each with its own regularities.  Predictive laws can only be applied when one has a “nomological machine”, meaning (so far as I can tell) an isolated system where all forces take one of the expected forms.  If one takes seriously the Aristotelian claim that the world is governed by natures rather than a fundamental law, arguably this is what one would expect.

Usually, when one imagines laws of physics breaking down, the idea is that they fail in some untested regime, such as very high energy.  Cartwright appears to be more interested in the idea that they would fail for untested functional forms.  For example, one’s dynamical laws may work for a quadratic but not a quartic potential.  It would be hard to rule out such a possibility, but none of the cases considered in the book pose particular difficulty for the “fundamentalist”, as Cartwright calls believers in unified laws.  Take her example of predicting where a thousand dollar bill carried by the wind will land.  Because we don’t know the exact initial conditions of bill or air, we should consider our initial state as a distribution in some vast state space evolving according to a sort of Liouville equation.  Such a calculation would correctly predict that the bill may fall anywhere in a wide area.  (A full calculation of this sort would be unfeasible, but a sense of the range of possible behavior could easily be gotten by numerically evolving the fluid+bill surface equations for, say, a dozen initial states within the range of uncertainty.)  Just because Newtonian mechanics is uninformative in such cases doesn’t make it wrong.  A more central example in the book is the BCS theory of superconductivity.  Cartwright makes a great deal of the the fact that the BCS Hamiltonian operator is a combination of what one might call “stock components”.  I agree that this illustrates well something about how modeling really works in physics, that in practice one does not start with the most general, fundamental action principle and then start approximating or dropping terms to get to one’s desired system.  However, the message is not that quantum mechanics has some problem with more general Hamiltonians, but that the goal is to capture the essence of superconductivity, so inessential and particularizing details have been dropped.  And it’s easy to see why certain components (esp. quadratic potentials) would be expected to appear in many contexts, as Cartwright acknowledges.  None of this is to say that she is wrong, but the fundamentalist position remains viable.  (Making it explicit and putting it into question, though, can only be good.)qw

Cartwright argues that the way scientists go about their business assumes an Aristotelian multitude of natural capacities and does not fit with Hume’s model of laws as a summary of observed regularities.  She points out that charges moving according to, say, the Coulomb electrostatic force never exactly happens.  The electrostatic force rather describes a tendency, a “what would have happened” if other effects had not been present.  The book then gives an intriguing discussion of Goethe’s criticisms of Newton, arguing that the success of Newton’s optical experiments is more consistent with an Aristotelian world of discrete natures than with Goethe’s more holistic world.

The book ends with some speculations on the relation between classical and quantum mechanics.  Cartwright proposes a sort of property dualism for the physical world, in which systems might have both quantum and classical properties, which presumably would have to interact somehow.