Outline Bacon’s ‘inductive methodology’ for arriving at the most real forms of simple natures. Is it undermined by Bacon’s own observation that heat is associated with ‘colliquation’ in wax but ‘induration’ in clay? Paul Mainwood June 2003

1

Introduction

Bacon’s Inductive Method for acquiring knowledge was designed with two principal aims in mind. First, he wished to avoid certain mistakes he recognised as recurring in human thought. These ‘Idols’ he carefully enumerated in the first book of the Novum Organum, discussing how each arises from weaknesses of the mind. His radical solution was to reject creative thought as an untrustworthy method for moving from observation to general principles, and replace it with a wholly mechanical procedure. Second, the Method was to discover knowledge that would be practically useful; although valuing abstract reasoning, Bacon was scathingly critical when so-called knowledge had no practical application. Most prized was power to infallibly induce a given effect in the most general circumstances; but to acquire the knowledge to do this would involve “building in the human understanding a true model of the world” (NO.I.124). For such an ambitious project, Bacon had to produce a radically original method. As he says, it would be “self-contradictory to expect that things which have never yet been done can be done except by means which have never yet been tried” (NO.I.6). At first sight, these original means may seem elusive - just like his predecessors he seeks out “Forms” of natures by means of “inductive” reasoning. But we are warned not to be misled by the terms he uses: To me it seemeth best to keep way with antiquity usque ad aras, and therefore to retain the ancient terms, though I sometimes alter the uses and definitions. (AL.VII.2) 1

So to understand Bacon’s Method, we must discover which terms he is using in accordance with tradition, and which he has altered for his own purposes. In §2 I shall discuss Bacon’s most important changes, bring out some substantive assumptions that are implied by the re-definitions, and in §3 show how they are required for the success of his Inductive Method. In §4 and §5, I shall argue that empirical evidence shows that these assumptions do not hold in the actual world; but if the Method is complemented with some corrective procedures - referred to, but not described in his surviving work - the assumptions may be weakened somewhat. However, on examination in §6 it is argued that no corrective procedure will save his Method if the assumptions are violated more than a little, and - unfortunately for Bacon - they are.

2

Definitions

Bacon’s Natural Science is divided into ‘Physic’, the study of material and efficient causes; and ‘Metaphysic’,1 the study of formal and efficient causes. He effectively discards one aspect of each - the material and final causes respectively - leaving Physic describing the specific, efficient events that cause a change, and Metaphysic looking for the universal Forms of natures that underlie them. However, Bacon used these terms in a very different way to Plato or Aristotle.

2.1

Natures

Nowhere does Bacon clearly define a “Nature” but he does give many examples. His lists are mainly of simple properties of bodies but he also recommends the use of aids to the senses such as microscopes and telescopes, and also includes quite complex phenomena such as the action of the tides and the motion of the moon. So natures appear to be phenomena of all kinds, which are usually - though not necessarily - directly observable. They are susceptible to investigation as to their causes, and Bacon’s ultimate aim is to subject all of them to human power. Bacon is most concerned with simple natures, a privileged subset that are particularly basic and widely distributed in the world. He assumes that there are some, such as heat or colours, that can be recognised relatively easily, although others may be misleading (NO.II.4). One of the Idols (of the Marketplace) is the mistake of using single terms to cover a variety of simple natures, or conversely, several terms for one. Assumption 1 At least some simple natures are easily identified as such. The simplicity of a nature depends on how the world is, not our state of knowledge (NO.II.5). 1I

follow Bacon in missing the ‘s’ from these terms, to emphasise their distinction from modern definitions.

2

Therefore, this is not merely a matter of definition, but a substantive assumption of our ability to identify an objectively distinguished class of phenomena.

2.2

Forms

The discovery of the “most real, simple Forms” is one of the two principal aims of Bacon’s new Method - the aim of human knowledge. It is inextricably linked to the other, which is to generate and superinduce new natures - the aim of human power. As he puts it, the roads to human knowledge and power “lie close together and are almost the same” (NO.II.4). Bacon is able to say this because he holds that once we have knowledge of a simple Form, it will immediately give us the power to produce the corresponding nature. For example, the simple natures of gold include yellowness, heaviness, malleability and lack of volatility, among others. Bacon thinks of gold as nothing more than a body with all of these natures, so if we know the Forms of each, we may join them all in one body, and thus produce gold (NO.II.4-5). Bacon characterises Forms as “disposing and leading to action”, but makes clear that they are stronger than dispositions as usually understood: The Form of a nature is such, that given the Form the nature infallibly follows. [...] Again, the Form is such, that if it be taken away the nature infallibly vanishes. (NO.II.4) So the presence of a Form is necessary and sufficient for the presence of its nature, which raises the question of how to distinguish between them. There are two aspects to Bacon’s characterisation of Forms. One is definitional: the Form is the nature expressed in more fundamental terms; the other closer to laws of nature: a Form possesses power to induce its nature in a body. Bacon tends to concentrate on one aspect at a time, leading to strong cases for interpreting Forms in each way. Many modern interpreters polarise to one view or another ([Horton(1973)] and [Quinton(1980)] mark the extremes of reading Forms as laws and as definitions, respectively.) Taking one quote in support of each: ... the Form of a thing is the very thing itself, and the thing differs from the form no otherwise than as the apparent differs from the real, or the external from the internal, or the thing in reference to man from the thing in reference to the universe. (NO.II.13) For when I speak of Forms, I mean nothing more than those laws and determinations of absolute actuality, which govern and constitute any simple nature, as heat, light, weight, in every kind of matter and subject that is susceptible of them. Thus the Form 3

of Heat or the Form of Light is the same thing as the Law of Heat or the Law of Light. (NO.II.17) Fortunately, in The Advancement of Learning, Bacon exposes the double definition as arising from confusing Metaphysic - the study of Forms as the essential definitions of natures; with Physic - the study of a nature’s material and efficient cause, which “carries” or “acts as the vehicle of” the Form (AL.VII.5). He gives the example of the whiteness of snow, which Physic shows to be caused by the subtle intermixture of air and water. If we contrast this with Valerius Terminus’ definition of the Form of whiteness: a mixture of substances in simple proportion, we can see how Forms have both definitional and law-like aspects. So Forms are indeed definitions of natures: they are an analysis of the nature to be explained in terms of other natures, and the discovery of such definitions is the aim of Metaphysic. But they are also laws: if we mix two substances in simple proportion, whiteness will always be produced. The particular substances we use and the way we mix them, “acts as the vehicle for the Form” and this vehicle is the subject of Physic. Bacon’s Method aims to find the most “true and real” Forms of simple natures, definitions in terms provided by “the universe and not the senses”. Bacon explains this as an analysis in terms of the most basic and general natures: The true Form is such that it deduces the given nature from some source of being which is inherent in more natures, and which is better known in the natural order of things than the Form itself. For a true and perfect axiom of knowledge then the direction and precept will be, that another nature be discovered which is convertible with the given nature, and yet is a limitation of a more general nature, as of a true and real genus. (NO.II.4) Putting these characteristics together, a true and real Form seems to be a scientific analysis of the more basic constituents of a nature, similar to the definition “Water is H2 O”, which tells us that the Form of water is the elements of hydrogen and oxygen in a particular molecular arrangement. Hydrogen and oxygen are more general and fundamental natures than water, as evinced by their also appearing as Forms of other natures (such as gaseous H and O2 ). However, a specific molecular arrangement and the 2:1 ratio “limits” these more general natures to the “true and real genus” of water. Of course, Bacon would not be content to rest with chemical elements, but would want to continue the analysis in terms of their constituent electrons and nuclei, until he reached the true, real, simple Forms, whatever they may be.2 2 Leaving

aside Bacon’s varying enthusiasm for aspects of Corpuscularianism (noting only that it would help

4

Assumption 2 There are a limited number of true, simple Forms in the world. Bacon draws an analogy with investigating a language, where we should not seek the infinite number of Forms of the sounds that make up the words, but those that make up the letters, a more tractable task. Nor again is there any reason to be alarmed at the subtlety of the investigation, as if it could not be disentangled; on the contrary, the nearer it approaches to simple natures, the easier and plainer will everything become; the business being transferred from the complicated to the simple ... as in the case of the letters of the alphabet and the notes of music. And inquiries into nature have the best result, when they begin with physics and end in mathematics. (NO.II.8)3 This assumption, that there is a finite “alphabet of the universe”, appears to be a simple article of faith for Bacon, part of his general trust in the comprehensibility of the world. (The final sentence is not followed up, but could be read as a conjecture that the ultimate simple natures may only be described mathematically.)

2.3

Latent Configurations and Processes

As well as a Form directly inducing a nature, there are circumstances in which a nature may be induced by a “Latent Process”. Bacon’s examples of these include the growth of plants and the generation of minerals (NO.II.5). They are “special habits of nature, not her fundamental and universal laws”. So although knowledge of them is useful, it will not lead us to the most general laws for inducing natures. For example, we may discover the specific process by which gold is produced in the earth’s crust, but unless we had access to exactly the same conditions, this would not help us make it. Bacon accordingly treats such knowledge as second-class, due to its limited scope in increasing human power. From what has been said, it is clear that the study of “Latent Configurations and Processes”4 must belong solely to Physic, and Bacon confirms this (NO.II.9). This is not to say that the natures involved are not ultimately governed by Forms, but studying these observable natures will not directly help to discover them, since the mechanism is latent, or hidden. At (NO.II.17), Bacon tells us that he is ignoring Latent Processes in the first part of his Inductive Method, and will return to them later, to see how they are found in concrete substances and natures. ground his assumptions). See [Harre(1964), pp.77-9]. 3 The same analogy also appears at (AL.VI.5), [Bacon(1857), IV:29], (NO.I.121), always in the context of the viability of his Method. Urbach’s denial that Bacon made such an assumption appears unsustainable. 4 Bacon’s distinction between configurations and processes is between the unobserved being in motion and at rest, and has been made obsolete by the discovery of universal molecular motion. I shall use the terms interchangeably.

5

Assumption 3 Forms are absolutely general and infallible rules for producing their natures, and are the only such rules. This has several important consequences. First, when true Forms are successfully induced in a body they will give rise to their simple nature, independently of any others present. If we excite the specific motion that is the Form of heat, then heat will be generated: ... without taking into account whether the body be elementary or subject to celestial influence; whether it be luminous or opaque; rare or dense; locally expanded or confined with the bounds of its first dimension; verging to dissolution or remaining in its original state; animal, vegetable or mineral, water, oil or air, or any other substance whatever susceptible of the above-mentioned motion. (NO.II.20) Second, it is only due to this assumption that discoveries made in Bacon’s Physic can have relevance to his Metaphysic. The only empirical marks of the Forms are their generality and infallibility, they are the fundamental laws of nature and admit of no exception. In contrast, a ‘law’ or rule based on a Latent Process would be descriptive, might include ceteris paribus clauses and would not be general or infallible. We can also see why it was with Metaphysic that Bacon was primarily concerned. Physic studies phenomena as they are tied to specific causal processes, whereas Metaphysic is concerned with general, exceptionless and universal principles about the physical world. As Bacon wanted to maximise man’s power to induce natures, he sought Forms, which gave rules that were “certain, free [= general5 ] and disposing to action.” (NO.II.4). Any specific or fallible method such as a Latent Process, was inferior to knowledge of a nature’s true Form.

3

The New Method of Induction

Like “Form”, the term “Induction” was used by Bacon in a totally different way to his predecessors (and indeed those who came after him.) His “new, certain method” was repeatedly contrasted with the old induction by enumeration which he discarded as “childish” and prone to error. The central differences are first, the eliminative character of his new method, proceeding from observation by ruling out natures that could be definitely shown not to be part of the Form.6 Second, the method was mechanical, replacing creative conjecture (which would be vulnerable to the Idols) with a rule-governed procedure. This he compared to drawing a circle with a compass 5 See

Valerius Terminus, especially (VT.11) for Bacon’s use of “free” in the context of his Method. gives many clear statements of the primacy of exclusion - eg: [Bacon(1857), IV:25], (NO.II.15 & 16)

6 Bacon

6

rather than freehand. Avoiding error with such aids is easy, and depends little on one’s natural ability (NO.I.61). Finally, Bacon wanted his method to proceed cautiously and step-by-step, not postulating conjectures too far ahead of the facts [Bacon(1857), IV:25]. He held that these features went a long way towards eliminating the Idols from reasoning. An notable feature of the Inductive Method is that it proceeds on the evidence of practical experiments, gaining knowledge only of specific efficient causes: the study of Physic. However, in Bacon’s taxonomy, its aims of discovering true, real Forms is a Metaphysical project. The eliminative method was designed to move from obvious but complex, efficient causes to the hidden but simple, formal causes. The idea seems to be that the general and infallible formal causes could be isolated, as the factors common to all the fallible efficient causes. We have only two examples of Bacon’s Method in action, in Valerius Terminus - applied to whiteness and in Novum Organum - applied to heat. The latter is far more developed, but still incomplete: Novum Organum - itself only the second book of the Great Instauration - finishes abruptly after the “First Vintage”, or “Indulgence of the Understanding”, which Bacon makes clear is only a preliminary stage, with much important work still to be done. In the Plan of the Great Instauration, we find that the existing section ends just before Bacon plans to “sift and examine” his preliminary notions [Bacon(1857), IV:25].

3.1

The Surviving Part of the Method

The first step in Bacon’s Method is to pick a nature to be investigated. This choice is clearly most important, since if we do not pick a simple nature, it will not necessarily have a true, simple Form to discover. Nevertheless, Bacon makes use of Assumption 1, and chooses heat. The next step is to construct a “Natural History” a fairly random and unsorted table of instances which possess the nature of heat. (NO.II.10-11). Then we construct a second table in parallel, with instances as close as possible to each of those in the first, except that the nature of heat is absent (NO.II.12). A third table, of instances in which the nature is present to differing degrees, is also compiled haphazardly, from a wide variety of sources (NO.II.13). The process of exclusion can now begin. The tables are compared to one another, and candidate natures for belonging to the form of heat are rejected if a “Negative Instance” can be found - a situation in which either they are not present and heat is observed, or they are present and heat is not observed. Bacon also introduces a new requirement for a Form (although it follows from his definitions) - that it should increase and decrease as its nature does. Therefore candidates are also rejected if they do not conform to this rule, by reference to the third table.

7

Bacon calls this first process of Exclusion the “Foundations of true Induction” - Induction itself only being complete when the true Form is found. But as yet, not even the process of exclusion is finished, since: “exclusion is evidently the rejection of simple natures; if we do not yet possess sound and true notions of simple natures, how can the process of exclusion be made accurate?” (NO.II.19). Nevertheless Bacon is happy to put forward ideas that may be incorrect, since truth “sooner comes from error than confusion.” He only objected to hypotheses (or axioms, as he misleadingly called them [Horton(1973), p.247]) when they were not appreciated as such, and retained even when refuted by experiment. In his example of a “First Vintage”, Bacon suggests that motion is the Form of heat. This is not to say that the Form of heat is any motion, but a particular variety (recall the “limitation to a true real genus” of §2.2). In this case, he makes various limitations to the motion being expansive, upward, rapid and belonging only to the small parts of the body (NO.II.20), each limitation being justified by reference to the tables. The next stage involves examination of “Prerogative Instances”, classes of phenomena that Bacon thinks may be especially useful for his Induction, of which he discusses twenty-seven. I shall return to a few of these in §5, but together they answer many of the criticisms that Bacon’s method was na¨ıve, and show him to be an acute critic of experimental design and practice.7 After they have been discussed, but before showing how they are to be applied to the First Vintage, the Novum Organum ends abruptly, at what must be less than half-way through.

3.2

Evaluation

If we put this rough idea of the eliminative procedure together with the assumptions identified in §2, we can see why Bacon might have made his famous claims for the inevitable success of his method.8 If we have selected a simple nature to investigate (Assumption 1), its Form is made up of the only basic natures infallibly associated with it (Assumption 3), and there are only a finite number of Forms as possible bases (Assumption 2), it seems reasonable that a process of elimination should eventually be successful. Trusting that the mechanical Method could avoid the Idols of the mind, Bacon was remarkably optimistic. With a sufficient quantity of observation - estimated as six times the volume of Pliny’s Natural History 9 - he thought the whole task of science would be achievable in a few years.10 7 See

[Urbach(1987)] and [Horton(1973)] for robust defences of Bacon. (NO. Preface), (NO.I.36-7) and many other remarks. 9 Letter to Jean Antoine Baranzano [Bacon(1857), VII:377] 10 [Urbach(1987), p.131] speculates that Bacon was exaggerating in an attempt to secure a grant from James I (to set up what would have been the world’s first research laboratory, based on his fictional New Atlantis.) 8 Eg:

8

4

Colliquation and Induration

Bacon’s observation that heat is associated with different natures in different circumstances appears to cast doubt on the assumptions on which his Method is based. Briefly stated, the problem is this: Let us imagine that we are investigating heat, and we are sure that it is a simple nature (Assumption 1). If we examine the effect of heat in wax then we note that it is infallibly associated with colliquation, so there is some sort of law-like connection between the two. If this connection were due to having some part of the Form in common, then we would expect colliquation to be associated with heat in all bodies (recall that Forms should give rise to their natures independently of the body in which they are instantiated - Assumption 3). However, heat is just as infallibly associated with induration in clay, and there seems nothing in common between colliquation and induration that could belong to the Form of heat. There appear to be two explanations within Bacon’s framework, each of which have unwelcome consequences for his Method: 1. We are mistaken about the simple natures. There is a distinct simple nature of heat, induration or colliquation, in wax and in clay. 2. There is a Latent Process in operation.

4.1

Option One - Compound Natures

In his example in the Novum Organum, Bacon assumed that heat was a simple nature, but earlier admitted that it is often difficult to distinguish one simple nature from a mixture of several. Of course if heat is divided into several simple natures (say heat-of-wax and heat-of-clay), then each could have their own Form: one associated with colliquation and another with induration. This explanation has two difficulties: it would conflict with Assumption 1, and with Bacon’s text (since he takes heat as a simple nature in his later work). But in the passage in The Advancement of Learning to which this question refers, Bacon does not make observations about heat, but about fire: Fire is the cause of induration, but respective to clay; fire is the cause of colliquation, but respective to wax; but fire is no constant cause either of induration or colliquation: so then the physical causes are but the efficient and the matter. (AL.VII.4) We can add this to (NO.II.20) where Bacon tells us: “our notion of fire is popular and of no use; being made up of the combination in any body of heat and brightness.” (By this he presumably

9

means “of no use for Metaphysic,” since investigation of a such a compound nature could not isolate a true Form.) Of course, Bacon could have experimented without the brightness of fire, and found that heat alone gives exactly the same results. So maybe heat also is not a simple nature, or alternatively colliquation or induration are not simple. Each of these possibilities could be granted by Bacon, though he must hope that he does not have to resort often to an explanation of this sort, or Assumption 2 might be threatened. Also, it must be possible to discover and correct these errors within the Method; Bacon categorises them as “Idols of the Marketplace,” so was certainly aware of their importance. §5 will look at incomplete parts of Bacon’s work for clues as to how this could be done.

4.2

Option Two - A Latent Process

Another possibility is that rather than Forms producing natures directly, a Latent Process is producing colliquation and induration. Though they and heat are all simple natures, with their own real forms, the Latent Configurations of wax and clay are sufficiently different that the application of heat gives different results for each substance. Unlike Forms, Latent Processes may produce results specific to the medium in which they are induced. Bacon might be more willing to resort to this explanation, he even hints at it in the quote from Advancement, by pointing out that he is only looking at Physical causes rather than Metaphysical Forms; but it brings out a crucial weakness in his Method. We can only place observable natures in the tables, and these are a very small subset of all the possible components to a Form. Just because we cannot observe a nature does not mean it is not present, yet Bacon remarks that a single negative instance is enough to reject a candidate for a Form (NO.II.18). The possibility of Latent Processes obviously make this elimination procedure highly fallible. The Forms involved are, by definition, hidden from our sense, so we will miss natures that are in fact present. This may lead us reject candidate natures that may in fact be part of the Forms. It seems implausible that Bacon was not aware of this difficulty, so why did he look upon his method as certain? On this basis Bacon has been accused of elementary error: defending induction as the only certain route to knowledge, but then putting forward a method which crucially involved retroductive inferences (explaining observations in terms of unobserved variables and structures).11 This is unfair, and ignores his redefinition of Induction, which he makes quite clear has a retroduct11 For example, [Whewell(1847/1967), pp.237-40] makes such an accusation, but shows respect for Bacon, defending him against accusations of simple Inductivism.

10

ive element. What is not clear is whether his claims of certainty were mere propaganda (as the estimates mentioned in §3.2 surely must have been). To answer this question, we must speculate on what may have come next in the Inductive Method. It seems plausible that the uncompleted sections of the Novum Organum would have added corrective procedures to the Method which could go some way towards solving the problems of unobservable natures.

5

The Missing Sections

Bacon gives a programme for the Novum Organum at (NO.II.21) which implies that seven of the nine parts must be missing. There is no detailed description of what would have appeared under such tantalising titles as “The Rectification of Induction” but it is clear that he had some corrective task in mind. There are several Axioms in Book 1 that suggest that the First Vintage conclusions must be subject to further experimental testing, to uncover Latent Processes, or disentangle Compound natures (e.g: NO.I.28, 103, 106 & 129). To find details of suitable experiments, we need look no further than the Prerogative Instances, which Bacon has yet to give a role within his Method. They have been ignored by many commentators. For example, in his General Preface to [Bacon(1857)] Ellis notes that while Prerogative Instances occupy most of the second book of the Novum Organum, they are unnecessary to understand Bacon’s Method. I strongly disagree with this dismissal. Indeed the wax/clay example could be an example of a Migratory Instance, one in which a small change in experimental circumstances causes the nature to appear or disappear. Although heat was applied in each case, by changing the medium we have banished colliquation, replacing it with induration. Bacon’s comments on this Instance are instructive; after recommending them for investigation since they provide a clear plan for practice, he warns they might lead to: ... a false opinion concerning the Form, drawn from a view of the efficient. But the efficient is always understood to be merely the vehicle that carries the Form. This is a danger however easily remedied by the process of exclusion legitimately conducted. (NO.II.23) It is a pity that Bacon does not make it clear how exclusion can “easily” remove the difficulty. However, the wax/clay experiment certainly suggests that an efficient Latent Process may be in operation (as suggested in §4.2). If we can discover the presence of Latent Processes by experi-

11

ments, then those parts of the Natural History which are suspected of involving them might be discarded from the tables, and the conclusions of the First Vintage corrected accordingly. Other Prerogative Instances might be used to correct “notions of simple natures”. Here the Instances of Divorce (NO.II.37), appear most relevant.12 The wax/clay example also could be interpreted as an instance of Divorce (the other option suggested in §4.1). Unfortunately, as Bacon laments, the Instance is no help in indicating what to do next. “They determine nothing, but simply notify the separability of one nature from another.” The only option is to start again, with new guesses at simple natures. Suggesting further details would be mere conjecture, but Bacon certainly imagined some correction of the First Vintage, and the above seems a reasonable reconstruction, given the material available. So although he has told us that a single contradictory instance excludes a nature from the Form of heat, there may be a stage of appeal, where these negative instances are examined to see if a Latent Process or compound nature is affecting the exclusion process. Whatever the exact procedure Bacon had in mind for using the Prerogative Instances, we will end up with a dilemma whenever a situation such the wax/clay example arises. As we have seen, it could be either a Migratory Instance, or of Divorce. Either we accept that the effect is due to a Latent Process, and discard certain preliminary results; or we change our ideas of simple natures and recommence the Method completely. The difficulty is to decide which. If Latent Processes or mistakes about simple natures are rare, isolated occurrences; a self-corrective Baconian method might just be viable. However, if these situations occur as a matter of course, then the Assumptions are badly threatened, and with them, the Method.

6

Specific Simple Natures or General Latent Processes?

Bacon hopes to distinguish between Forms and Latent Processes due to the generality of their presence in the causes observed, but the wax/clay example shows a grey area. Whenever we apply heat to wax or to clay, we infallibly and generally induce colliquation or induration respectively. Either we are dealing with Latent Processes which are oddly general in that they are present in all wax and in all clay, or we have oddly specific Forms, with simple natures that are confined to wax and to clay. Further empirical investigation will not help; since generality is the only empirical difference between natures induced by a Latent Process and those by a Form, the choice will just recur. 12 Instances

of Companionship and the Fingerpost (NO.II.33 & 36) might also be thought to have some bearing.

12

For example, we might next discover that there is a particular variety of wax that solidifies when heated, and be back in the same situation. Unfortunately, the dilemma is not specific to the wax/clay example: if we weaken either Assumption 1 or 3, then we lose the only empirical differences between a nature produced directly by inducing its Form in matter, and one produced by a Latent Process underlying it. Note that the difference between the two situations does not exist for Physic13 which merely attempts to describe any regularities we come across (mathematically or otherwise14 ). It is up to Metaphysic whether we are studying Latent Processes or studying Forms, and in the absence of a clear line between general infallible laws and specific fallible rules there is no further empirical evidence to appeal to. But it is vital for Bacon to be able to draw this distinction, since his eliminative method is designed to find true Forms, it must be able to discard the interfering Latent Processes. He held that the first step of science is the establishment of the fundamental true Forms, only then proceeding to understand more complex processes (NO.II.7). Bacon’s Method always searches directly for a matter theory; attempting to analyse everything down to its ultimate components, and only then investigating the process by which those components give rise to the nature. Yet it has turned out that Latent Processes give rise to extremely powerful and general theories - think of thermodynamics, which can be codified into general systems of laws, though it makes no claim to have identified the ultimate components of matter. Theories as disparate as solid state physics, Newtonian mechanics and condensed matter also abstract from ultimate components; so according to Bacon’s taxonomy study Latent Processes. Einstein famously drew a distinction between Principle and Constructive theories (identifying Relativity with the former). Principle theories start with a few basic axioms and proceed to work out their implications, making no attempt to decide whether a Latent Process or Form is behind the phenomena. Constructive theories aim to provide concrete models of the reality underlying the phenomena. Clearly Bacon was only interested in Constructive theories, and would dismiss the “logical perfection and security of the foundations” of Principle theories [Einstein(1954/1973), p.229] as Idols of the Theatre (NO.I.56-65). But Principle theories must come first. Even with all the advances since Bacon’s time we cannot yet identify a single simple nature, let alone its Form. Unfortunately for Bacon his Assumptions simply do not hold in the actual world, so his method became irrelevant. Instead, Physic outstripped Metaphysic by discovering general laws - which all 13 Or

for Physics, as it actually developed. did value mathematics as a tool for investigation, although subsidiary to the aim of discovering true

14 Bacon

forms.

13

(or almost all) have turned out to be descriptions of Latent Processes. Nevertheless, the power they have given mankind to induce natures would not have disappointed Bacon. Word count: 5134

References [Bacon(1857)] F. Bacon.

The Works of Francis Bacon.

Fommann-Holzboog, Stuttgart-Bad

Cannstatt, 1963 facsimile edition, 1857. Translated by Robert Leslie Ellis, Douglas Heath and James Spedding.15 [Einstein(1954/1973)] A. Einstein. What is the theory of relativity? In C. Seelig and S. Bargmann, editors, Ideas and Opinions. Souvenir Press, London, 1954/1973. [Harre(1964)] R. Harre. Matter and Method. Ridgeview, Reseda, California, 1964. [Horton(1973)] M. Horton. In defence of Francis Bacon. Studies in the History and Philosophy of Science, 4(3):241–278, 1973. [Quinton(1980)] A. Quinton. Francis Bacon. Oxford University Press, 1980. [Urbach(1987)] P. Urbach. Francis Bacon’s philosophy of science : an account and reappraisal. Open Court, La Salle, Illinois, 1987. [Whewell(1847/1967)] W. Whewell. The Philosophy of the Inductive Sciences, volume 2. Johnson Reprint, New York, 2nd edition, 1847/1967.

15 In quoting, I have used VT for Valerius Terminus (translated from an unpublished manuscript of uncertain date, although the figures 1603 appear on the blank, along with some astrological symbols that support this), AL for the 1605 Advancement of Learning and NO for the the 1620 Novum Organum. References are then given by Book/Section number and Bacon’s aphorism number.

14