Alberto Vanzo writes…
On this blog, we have often stressed the importance of the movement of experimental philosophy between the 1660s and ca. 1800. What happened to this movement in the nineteenth century? Last week Juan noted that, after 1823, the entry “experimental philosophy” in the Encyclopedia Britannica was substantially shortened and then removed. This suggests that that notion may have disappeared from the British philosophical scene. Something similar had happened in Germany, where the tradition of experimental or, as it was mostly called, observational philosophy was eclipsed by Kantian and post-Kantian systems around 1800. The expression “observational philosophy” seems to have disappeared from the German philosophical vocabulary in the early nineteenth century.
In this post, I will highlight an exception to this trend: Pierre-Claude-François Daunou’s Recherches sur les systèmes philosophiques applicables à l’historie. This is the text of the lectures on the history of philosophy that Daunou gave at the Collège de France in 1829-1830. It was published posthumously in 1849. In this work, as late as in the mid-nineteenth century, we find a wholehearted defense of experimental philosophy and its application to philosophical historiography.
Daunou aims to outline a philosophical history of philosophy. What renders the history of philosophy philosophical? Daunou’s answer is: “history is philosophy, when it consists in a methodical series of facts that are carefully verified and presented as experimental instructions”. In fact, “only the experimental school provides the true method in the historical studies”.
Like Degérando before him, Daunou draws on the historical facts to develop a natural history of philosophy: a classification of the various types of philosophical systems which we can use to establish which is the best. To this end, “the classifications must resemble those of naturalists, that is, they must only summarize the facts. Pretending that they are given and established a priori by the nature of things is a Platonic illusion, that has introduced many prejudices and errors into the sciences”.
In 1829, Daunou could chose between plenty of alternative classifications: for instance, the old division of philosophers into sects to be found in Brucker’s manual, praised by Daunou; the empiricism/rationalism distinction used by the Kantians and by Degérando; and Cousin’s fourfold distinction between idealism, sensualism, scepticism and mysticism. Rejecting all of these classifications, Daunou follows Diderot, Condillac and Condorcet in relying on the good old division between experimental and speculative (or in his terms, contemplative) systems. On the one hand, we have the experimental approach of Aristotle, Bacon, Gassendi, Locke, and Condillac. On the other hand, we have the Platonic attempt to develop philosophical systems a priori. Kant, far from synthesizing these two trends as his disciples claimed, was responsible for perpetuating the Platonic, contemplative illusion. He “delayed the progress of science” and he induced French thinkers to accept the mistaken principle that “the abstract precedes the concrete, sheds light on it and dominates upon it”.
Given Daunou’s assumptions, it is easy to guess what moral he draws from his history of philosophy: we must abandon the Platonic “picture of an idea or imaginary world” and acknowledge that “we owe all progress of physical and moral sciences” to experimental philosophers.
What is surprising, or at least interesting, is that we find these claims in a text published as late as in 1849. Was Daunou a historian attardé, a living fossil in his own time, as Gregorio Piaia states in his very informative survey of Daunou’s work (to which this post owes much) in the Storia delle storie generali della filosofia? It is hard to deny that he was, at least to some extent. However, Daunou’s speculative-experimental distinction was paralleled in Saint-Simon’s contrast between Plato’s and Descartes’ vague speculations on the one hand, Aristotle’s and Bacon’s positive philosophy on the other. The term used by Saint-Simon is “positive”, not “experimental”. However, Saint-Simon’s positive philosophy was based on the experimental method. And in the first volume of Comte’s Course of Positive Philosophy we find the same contrast between the metaphysical spirit, to be rejected, and the positive spirit of those that Daunou regarded as experimental philosophers.
Let’s go back to our initial question on the fate of experimental philosophy in the nineteenth century. Daunou’s work, together with Saint-Simon’s and Comte’s statements, suggests that the notion of experimental philosophy was not simply abandoned in nineteenth century France. Instead, it morphed into the new notion of positive philosophy, or at least it contributed to the definition of this new important movement in the French philosophical scene. I am no expert in nineteenth century French philosophy though. I would love to hear if you find this suggestion plausible.
Juan Gomez writes…
Some time ago I wrote a post regarding David Fordyce’s Elements. This text was used almost in its entirety as the entry for moral philosophy in the Encyclopædia Britannica from the first edition in 1771 and until the seventh edition where it was modified and the replaced by an essay on the topic by William Alexander. I want to refer to the Encyclopædia again, this time to trace the description of four terms, namely ‘empiric,’ ‘experimental philosophy,’ ‘rational’ and ‘rationalism,’ and ‘speculative.’ I will focus on the first two terms in today’s post.
The word ‘empiric’ appears in the first eight editions (1771 to 1898 when the ninth edition appeared). It is a very short entry and it restricts its use to method in medicine:
It is clear from the definition that the word had a different use than the one implied by the modern term ‘empiricism,’ which appeared for the first time in the eleventh edition(1910) of the Encyclopædia. In such editions the writer of the entry tells us that the term refers “in philosophy, [to] the theory that all knowledge is derived from sense-given data. It is opposed to all forms of intuitionalism, and holds that the mind is originally an absolute blank.” The last paragraph of the entry refers to the restricted definition of the term ‘empiric’ given in previous editions that I quoted above.
This term can be found in the first eight editions as well and disappears from the ninth edition onwards. The following is the definition from the first edition of the Encyclopædia:
However, the definition for Experimental Philosophy was substantially expanded for editions two to six, and then reduced to one small paragraph in the eighth edition. From 1778 (second edition) to 1823 (sixth edition) the entry consists in a general description and refers to seventeen items that form “the foundations of the present system of experimental philosophy.” The items are basic definitions of the object of study of experimental philosophy: natural bodies and their properties, extension, arrangement of particles, law of gravity, properties of light, and so on. For the seventh edition the entry was reduced to this:
The entry is the same for the eighth edition but from the ninth edition onwards there is no entry for ‘experimental philosophy.’
As far as the Encyclopædia Britannica is concerned, we can see that the terms used in the eighteenth and nineteenth centuries are better explained by using the ESD framework instead of the RED. The contrast between the meaning of the term ’empiric’ in the medical context and the later twentieth century entry on ’empiricism’ illustrates this nicely. As we will see in my next post, the way the terms ‘speculative’ and ‘rational’ were used gives us more evidence to prefer the ESD framework for interpreting the early modern period.
Peter Anstey writes…
In a recent article Peter Harrison has drawn our attention to the phenomenon of experimental Christianity in seventeenth-century England (‘Experimental religion and experimental science in early modern England’, Intellectual History Review, 21 (2011)). In this post I would like to take up where Harrison left off and discuss one proponent of experimental religion whom Harrison does not mention, namely Francis Bampfield (1614–1684). Bampfield provides an interesting case study because while he was a promoter of experimental Christianity, he was also a harsh critic of the new experimental natural philosophy.
In two works from 1677, All in One and SABBATIKH, Bampfield lays out his case against experimental natural philosophy. In his view the source of all useful and certain knowledge is the Scriptures.
Practical Christianity, and experimental Religion is the highest Science, and the noblest Art, and the most honourable Profession, which gives light to all inferiour knowledges, and would admit into the Royalest Society, and draw nearest in resemblance and conformity, to the glorified Fellowship in the Heavenly College above, where their knowledge is perfected in visional intuitive light. Here is the prime Truth, the original Verity, as to the manifestativeness of it in legible visible ingravings, which would carry progressively into other Learning contained therein: all here is reducible to practice and use, to life and conversation; here existences and realilties are contemplated and proved, not mere Ideas and conceits speculated as elsewhere. (All in One, p. 12)
It is the mere ideas and conceits of the new experimental philosophy of the Royal Society, or the Fellows of Gresham College, that Bampfield is concerned to expose:
How many thousands have by their wandring after such misguiders left and lost their way in the dark, where their Souls have been filled with troublesome doubts, and with tormenting fears, exposing them to violent temptations of Atheism and Unbelief? and what wonder, that it is thus with the Scholar, when some of the learnedest of the Masters themselves have resolved upon this, as the conclusion of all their knowledge, that, All things are matter of doubtful questionings, and are intricated with knotty difficulties, and do pass into amazing uncertainties, and resolve into cosmical suspicions? And this, not only is the deliberate Judgement of particular Virtuoso’s in our day, but has been the publick determination of an whole University. (All in one, p. 3)
What are these ‘knotty difficulties’ that pass into ‘amazing uncertainties’ resolving into ‘cosmical suspicions’? The alert reader will no doubt see here a direct allusion to Robert Boyle’s Tracts of 1670 in which he discusses cosmical qualities that seem to have ‘such a degree of probability, as is want to be thought sufficient to Physicall Discourses’ (Works of Robert Boyle, eds Hunter and Davis, London, 1999–2000, 6, p. 303). Boyle appended to his essay on cosmical qualities another on cosmical suspicions which contains just the sort of speculative reflections that Bampfield is alluding to here. (As far as I can determine, Boyle’s is the first work in English that uses the term ‘cosmical suspicions’.) That Bampfield was a close reader of Boyle’s writings comes out in a later passage which I quote in extenso:
There is an honourable Virtuosus, who has travelled far in Natures way, and has made some of the deepest inquiries into Experimental, Corpuscular, or Mechanical Philosophy, that in the requisites of a good Hypothesis amongst others of them, doth make this to be one of its conditions, that it fairly comport not only with all other truths, but with all other Phaenomena of Nature, as well as those ’tis fram’d to explicate, and that, not only none of the Phaenomena of Nature, which are already taken notice of do contradict it at the present, but that, no Phaenomena that may be hereafter discovered, shall do it for the future. Let it therefore from hence be considered, whether seeing, that History of Nature, which is but of human indagation and compiling, is so incomplete and uncertain, and many things may be discovered in after-times by industry, or in some other way by providential dispensing, which are not now so much as dreamed of, and which may yet overthrow Doctrines speciously enough accommodated to the Observations, that have been hitherto made (as is by himself fore-seen and acknowledged) whether now, the only prevention and remedy in this case (which is otherwise so full of just fears, of real doubts, of endless dissatisfaction, and of perplexing difficulties) be not, to bring all sorts of necessary knowledges to the Pan-sophie, the Alness of Wisdom, in the Scriptures of Truth, where none of the forementioned Scriptures have any ground to set their foot on, in regard that Word-Revelations about Natures Secrets, are the unerring products of infinite Wisdom, and of universal fore seeingness, which are always uniform and the same, in their well-established order, and stated ordinary course without any variation, by an unchangeable Law of the All-knowing Truthful Creator, and Governour, and Redeemer. (All in One, pp. 56-7)
Bampfield is, of course, referring to Boyle’s Excellency of Theology (Works, 8, p. 89) and while he is cautious not to be overtly critical of Boyle here, the thrust of his comments is to undermine the epistemic status of the experimental philosophy, calling it ‘incomplete and uncertain’. For, as he says in his sequel SABBATIKH:
the unscriptural way they take in their researches into natural Histories and experimental Philosophy, will never so attain its useful end for the true advance of profitable Learning, till more studied in the Book of Scriptures, and suiting all experiments unto this word-knowledge. (SABBATIKH, p. 53)
It is ‘word knowledge’ and not knowledge of the world that Bampfield is defending. What Boyle himself made of all of this, if it even came to his attention, we will never know. He never mentions Bampfield in any of his works or correspondence.
Kirsten Walsh writes…
In the first edition of Principia (1687), book 3 contained nine hypotheses. But in the second edition (1713), Newton re-structured book 3 so that it contained only two hypotheses. All but one of the old hypotheses were simply re-labelled. Those that specified explanatory constraints were called ‘Rules of Reasoning’, those that were simply unsupported generalisations were called ‘Phenomena’, and only the assumptions about nature were called ‘Hypotheses’.
|1st edition||2nd edition|
|Hypothesis 1||Rule 1|
|Hypothesis 2||Rule 2|
|Hypothesis 3||Replaced by Rule 3|
|Hypothesis 4||Hypothesis 1|
|Hypothesis 5||Phenomenon 1|
|Hypothesis 6||Phenomenon 3|
|Hypothesis 7||Phenomenon 4|
|Hypothesis 8||Phenomenon 5|
|Hypothesis 9||Phenomenon 6|
|Lemma 3||Hypothesis 2|
Table: Book 3 changes from 1st to 2nd editions.
These changes seem to indicate that Newton’s attitude to hypotheses changed dramatically between 1687 and 1713 – probably in response to Leibnizian criticisms and Cotes’ editorial comments.
On this blog, I have given you many reasons to suppose that, as early as 1672, Newton was working with a clear epistemic distinction between theories and hypotheses. More recently, I have argued that in fact, Newton was working with a three-way epistemic distinction between theories, which are certain and experimentally confirmed, hypotheses, which are uncertain and speculative, and queries, which are not certain, but provide the proper means to establish the certainty of theories. I call this division Newton’s ‘epistemic triad’. I argued that hypotheses perform a distinctive and vital supporting role to theories and queries, and that this role is an enduring feature of Newton’s methodology.
Today I’ll compare the roles of hypotheses and rules of reasoning in book 3, and argue that Newton’s attitude to hypotheses c.1713 was a refinement of his attitude c.1687, but not a dramatic change.
To begin, consider hypothesis 1 (2nd edition): “The centre of the system of the world is at rest.”
Upon introducing this hypothesis, Newton explained that:
- “No one doubts this, although some argue that the earth, others that the sun, is at rest in the centre of the system. Let us see what follows from this hypothesis.”
This is a simplifying assumption. From this hypothesis, in conjunction with Corollary 4 of the Laws of Motion,
- “The common centre of gravity of two or more bodies does not change its state whether of motion or of rest as a result of the actions of the bodies upon one another; and therefore the common centre of gravity of all bodies acting upon one another (excluding external actions and impediments) either is at rest or moves uniformly straight forward”,
Newton derived Proposition 11: “The common centre of gravity of the earth, the sun, and all the planets is at rest.”
This enabled him to calculate the motions of the planets – that is, to deduce the observational consequences of his theory. I consider this to be the chief role of hypotheses, and it is experimental.
Now compare this with how Newton uses his rules of reasoning.
Rule 1 states: “No more causes of natural things should be admitted than are both true and sufficient to explain their phenomena.”
And Rule 2 states: “Therefore, the causes assigned to natural effects of the same kind must be, so far as possible, the same.”
In his discussion of Proposition 4, Newton explained:
- “And therefore that force by which the moon is kept in its orbit, in descending from the moon’s orbit to the surface of the earth, comes out equal to the force of gravity here on earth, and so (by rules 1 and 2) is that very force which we generally call gravity.”
These rules didn’t help Newton to deduce the observational consequences of his theory in the same way as his hypothesis 1. They provide an important supporting role for his theory, but it is not an experimental role.
To the second edition, Newton also added the General Scholium, in which he (in)famously declared “hypotheses non fingo”. Recently I argued that, in this passage, Newton was not railing against hypotheses in general, but rather, against the use of ‘causal hypotheses’ to illustrate more abstract theories. Given that the first edition did not contain any causal hypotheses, I consider this addition to indicate Newton’s increasing conviction in his method, rather than any dramatic change.
So finally, to summarise the developments between the first and second editions of Principia:
In the first edition:
- Hypotheses are temporarily assumed, untestable propositions that provide a supportive role; and
- There are no causal hypotheses.
In the second edition:
- Hypotheses are temporarily assumed, untestable propositions that provide a supportive experimental role;
- Other temporarily assumed, untestable propositions are given other labels to distinguish them from hypotheses; and
- Emphatically, there are no causal hypotheses.
I see these changes as developments of Newton’s epistemic triad, rather than dramatic methodological changes.
Alberto Vanzo writes…
As we have already noted on this blog, “empiricism” was mainly used as a medical term in the early modern period. It referred to empirical physicians, a movement that had its origins in ancient Greece. Greek empirical physicians claimed that our knowledge of medical cures derives entirely from experience and derives its justification from experience. It does not derive from any insight into “hidden natures, causes, and actions, not open to observation, but only accessible to reason, e.g., atoms, invisible pores, functions of organs, or essences” (Michael Frede).
These views of ancient empirical physicians were remarkably close to those of the early modern physicians who were associated with the experimental philosophy. They too rejected reasonings from principles and speculations on hidden essences and claimed to derive their cures from experience alone. The early moderns were conscious of this similarity between ancient empirical physicians and modern experimental physicians. They acknowledged that, since empirical physicians relied on “experiments” (that is, experience), their teachings contain “something that is certaine & experimentall” (Hartlib Papers).
In the light of this, it is puzzling that the upholders of the “experimentall” approach to medicine used to criticize medical empiricism and empirical physicians. They generally accepted the then common association of empirical physicians with “Mountebanks, pretended Chymists, Apothecaries, Chirurgeons, Midwives, &c. in which piece of Folly the English surpass all the Nations of Christendom.”
To be sure, there were some cases in which early modern experimental physicians praised ancient medical empiricism, portraying it as a forerunner of experimental medicine. But as far as I know, these were rare exceptions. Why were they exceptions and not the rule? Why did experimental physicians see themselves as opponents, rather than followers, of medical empiricism?
I have found some answers to these questions in the writings of two prominent eighteenth-century physicians. They are the German physician Friedrich Hoffman, a leading systematist of the early eighteenth-century, and the Scottish physician John Gregory, who taught at Edinburgh in the latter half of the century. They both associate medical empiricism with the application of medicines that happened to work in past cases to patients which display similar symptoms. This approach was epitomized in the “empirical books” that were compilations of medical recipes and cures for all sorts of diseases. According to Gregory, this approach raises two problems.
Firstly, it relies on experience in the wrong way. Empirical physicians rely on the recipes that have been shown to work by experience, but true physicians should also rely on experience to improve their remedies and practice.
Secondly and more interestingly, empirical physicians did not pay sufficient attention to experience. They did not carefully inquire into the circumstances of individual patients. Instead, having noted few symptoms, they hastily prescribed a familiar remedy. As Hoffmann pointed out, this “obtuse and most dangerous empiria” was deeply flawed because the “efficacy of a cure” does not reside in the nature of a remedy, but in the way in which the remedy interacts with the specific circumstances of a patient. In Gregory’s words, since “empirics” paid insufficient attention to those circumstances, “notwithstanding their pretensions of relying upon experience alone, have in truth abandoned it”.
Gregory’s preferred alternative to this flawed medical empiricism was what appears to be, broadly speaking, an application of the Newtonian experimental method to medicine. Gregory praised a medical theory that “is produced by practice, is founded on facts alone, and constantly appeals to them for its truth”. He rejected the use of hypotheses, considered as principles devoid of “proof from experience”. He advocated their use, as long as they were “proposed in the modest and diffident manner that becomes mere suppositions or conjectures”, that is, queries. Used in this way, hypotheses “are not only harmless but even necessary in establishing a just theory in medicine”. Gregory portrayed this experimental method as an alternative to medical empiricism. Only one century later, after “empiricism” took on a new meaning, did it come to be regarded as a positive stance even within medicine.