\documentclass[% version=last,% a5paper, 12pt,% headings=small,% twoside,% titlepage=false,% open=any,% cleardoublepage=empty,% bibliography=totoc,% %parskip=half,% draft=true,% %DIV=classic,% same as 10 under 10pt type DIV=12,% headinclude=false,% pagesize] {scrbook} %\usepackage[nocite]{showlabels} \usepackage{scrpage2} \pagestyle{scrheadings} \clearscrheadings \ofoot{\pagemark} \setkomafont{pageheadfoot}{\normalfont} %\rehead{\emph{Essay on Metaphysics}, Part III\textsc c: Causation} \lofoot{\leftmark} \usepackage{moredefs,lips} \usepackage[normalem]{ulem} \usepackage{verbatim} %\let\texdots\dots %\renewcommand{\dots}{ . . . } \newcommand{\mychap}[2][]{\addchap[#1]{#2}\stepcounter{chapter}} \renewcommand{\raggedsection}{\centering} \usepackage{makeidx} \makeindex \deffootnote[1em]{1em}{1em} {\textsuperscript{\thefootnotemark}} %\usepackage[ruled]{manyfoot} \usepackage[ruled]{bigfoot} \usepackage{perpage} %\SetFootnoteHook{\noindent} % This didn't do anything \DeclareNewFootnote{A}[fnsymbol] % Collingwood's notes \MakePerPage{footnoteA} \DeclareNewFootnote{B}[arabic] % my notes on the ideas %\MakeSorted{footnoteB} \DeclareNewFootnote{C}[alph] % my notes on typography % the bracketed [alph] is not necessary if the following is used: %\renewcommand{\thefootnoteC}{\textbf{\textit{\alph{footnoteC}}}} \renewcommand{\thefootnoteC}{\textit{\alph{footnoteC}}} \DeclareNewFootnote{D}[roman] % notes on notes \MakePerPage{footnoteD} %\usepackage{chngcntr} %\counterwithin*{footnote}{chapter} %\counterwithin*{footnoteB}{chapter} %\counterwithin*{footnoteC}{chapter} \usepackage{relsize} \renewenvironment{quote}{\begin{list}{} {\relscale{.90}\setlength{\leftmargin}{0.05\textwidth} \setlength{\rightmargin}{\leftmargin}} \item\relax} {\end{list}} \renewenvironment{quotation}{\begin{list}{}% {\relscale{.90}\setlength{\leftmargin}{0.05\textwidth} \setlength{\rightmargin}{\leftmargin} \setlength{\listparindent}{1em} \setlength{\itemindent}{\listparindent} \setlength{\parsep}{\parskip}} \item\relax} {\end{list}} \usepackage[polutonikogreek,english]{babel} \usepackage{gfsporson} \newcommand{\gr}[1]{\foreignlanguage{polutonikogreek}{\relscale{0.9}\textporson{#1}}} \usepackage{hfoldsty,url} \usepackage{textcomp} % \copyright did not have a circle without this % None of the following are as good as Computer Modern; % they are usually too heavy %\usepackage{mathptmx} %\usepackage{mathpazo} %\usepackage{charter} %\usepackage{utopia} %\usepackage{chancery} %\usepackage{bookman} %\usepackage{newcent} \usepackage[neverdecrease]{paralist} \renewcommand{\vec}[1]{\mathbf{#1}} \begin{document} \title{Causation} \subtitle{Part III\textsc c of \emph{An Essay on Metaphysics}} \author{R. G. Collingwood} \date{1940} \publishers{Edited and annotated by\\ David Pierce\\ Mathematics Department\\ Mimar Sinan Fine Arts University\\ Istanbul\\ \url{mat.msgsu.edu.tr/~dpierce/}\\ \url{polytropy.com}\\ November 10, 2014\\ Corrected January 26, 2015\\ Edited March 28, 2019} \maketitle \tableofcontents \refoot{On Collingwood's ``Causation''} \mychap[Preface]{Preface} R. G. Collingwood's \emph{Essay on Metaphysics} (1940\nocite{Collingwood-EM1}) consists of three parts: \begin{compactenum}[(I)] \item Metaphysics, \item Anti-Metaphysics, \item Examples. \end{compactenum} The last part itself consists of three parts: \begin{compactenum}[(A)] \item The Existence of God, \item The Metaphysics of Kant, \item Causation. \end{compactenum} The present document consists the six chapters of Part C of Part III, along with some notes of my own, written in the ongoing process of understanding Collingwood's text. I do this work as an \emph{amateur} of philosophy. My formal connection to the subject comes only through completing the undergraduate ``New Program'' at St John's College \cite{Pierce-SJC}. I also have the degree called Doctor of Philosophy; but the degree is actually in mathematics. The Introduction of the present document is one of the notes just mentioned. The chapters of the Appendix are more notes. Otherwise, my notes are footnotes, of three kinds: \begin{compactenum}[1)] \item on ideas, numbered consecutively throughout the document by Arabic numerals: 1, 2, 3, 4, \lips; \item on typography, numbered consecutively by italic minuscule Latin letters: \textit{a, b, c, d,} \lips; \item on notes themselves, numbered by minuscule Roman numerals: i, ii, iii, iv, \lips%%%%% \footnoteC{For the multiple footnote sequences, I use the \url{bigfoot} package for \LaTeX, based on the \url{manyfoot} package. The latter is documented as part of the bundle called \url{ncctools}. For reasons unknown to me, footnotes can be needlessly split across two pages.%%% \footnoteD{There can be other difficulties, as when a footnote that \emph{must} be broken between pages seems not to be broken in the best place. This may have to do with the normal page-breaking algorithm implemented by the \url{KOMA-script} document class that I use.} %%%%%%%%%%%%%%%%%%%%%%%%%% Without \url{bigfoot}, the \LaTeX\ default is to number footnotes by chapter. With \url{bigfoot}, this does not happen; if one wants it to happen, one can use the commands of the \url{chngcntr} package. I do not find these matters discussed in the \url{bigfoot} or \url{manyfoot} documentation.} %%%%%%%%%%%%%%%%%%%%%% \end{compactenum} Collingwood's own footnotes are marked now by symbols%%%%%%%%%%%%%%%%%% \newcounter{illustration}\stepcounter{illustration} (\fnsymbol{illustration}\stepcounter{illustration} %\fnsymbol{illustration}\stepcounter{illustration}, %\fnsymbol{illustration}\stepcounter{illustration}, %\fnsymbol{illustration}\stepcounter{illustration}, %\fnsymbol{illustration}\stepcounter{illustration}, %\fnsymbol{illustration}\stepcounter{illustration}, %\fnsymbol{illustration}\stepcounter{illustration}, and \fnsymbol{illustration}\stepcounter{illustration})---% there are nine in all, and originally they were marked by Arabic numerals, starting with 1 on each page (and only one page had a second footnote). A revised edition of the \emph{Essay on Metaphysics}, ``with an Introduction and additional material edited by Rex Martin,'' was published by Oxford with the following notice: %%%%%%%%%%%%%%%%%%%%%%% \begin{center}\smaller First edition \copyright\ Clarendon Press 1940\\ Revised edition \copyright\ Teresa Smith 1998; introduction and new\\ annotation \copyright\ Rex Martin 1998 \end{center} Teresa Smith is Collingwood's daughter. I possess the revised edition of the \emph{Essay} in the paperback version published in 2002 \cite{Collingwood-EM}. The editor's Preface reports, ``the original text\lips has been left completely unchanged, including even the pagination.'' Those original page numbers are bracketed and bolded in the transcription below, which is of the original pages 285--343.%%%%%%%%%%%% \footnoteC{Each chapter of the original text begins on a new page, and a number is not printed on this page; neither then is its number given in the transcription.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% I make no use of these numbers though; any references made by me to passages of the present text use the present pagination. When Collingwood himself refers to such a passage, as on pages \pageref{C-int-ref} and \pageref{C-int-ref-2}, I have added in brackets the page of the present edition. There are also some references, by Collingwood and by me, to pages of the \emph{Essay} that are not in part III\textsc c; these can only use the original pagination, which might overlap with the pagination of the present document. My comments on Collingwood's work, in the footnotes numbered by Arabic numerals, were originally heavier in the earlier chapters of the text, because here is where the \emph{moral} implications are stronger, and these implications are my main interest. Since the original draft of this document though, I have spent some time looking back at Newton's \emph{Principia} because of Collingwood's discussion in Chapter XXXII. For Chapter XXXIII, one should consider Kant more carefully than I have done so far. In Chapter XXXIV, Collingwood cites several philosophers of his day, but I have not looked up their work for myself. Their misunderstanding of metaphysics and its relation to science may be a moral error, but I am not sure how important their specific error is today. We have other errors or misunderstandings involving science. These are themes of Robert Pirsig's \emph{Zen and the Art of Motorcycle Maintenance} \cite{ZAMM} and Mary Midgley's \emph{Evolution as a Religion} \cite{Midgley-ev}, to name two books I know and like. I am interested in questions of \emph{responsibility} as raised by Collingwood's consideration of cause. Bill Watterson considers it in the four panels of \emph{Calvin and Hobbes} for February 15, 1995 \cite[Book Three, page 340]{Watterson}, where the first title character watches television and remarks: \begin{quote}\scshape \begin{asparaenum}[1.] \item Graphic violence in the media. \item Does it glamorize violence? Sure. Does it desensitize us to violence? Of course. Does it help us tolerate violence? You bet. Does it stunt our empathy for our fellow beings? Heck yes. \item Does it \textbf{cause} violence? \lips Well, that's hard to prove. \item The trick is to ask the right question. \end{asparaenum} \end{quote} Perhaps the notion of ``cause'' used here (and emphasized in the original) is influenced by science. More precisely, it may be influenced by a conception of science that is either obsolete or ill understood. There are more examples in some of the notes. \mychap[Introduction]{Introduction} %\mychap[Summary]{Summary} The mode of inquiry that gives a name to Aristotle's \emph{Metaphysics} is the \textbf{historical science of absolute presuppositions.} So argues Collingwood in \emph{An Essay on Metaphysics.} The part of the \emph{Essay} on causation is an example of doing metaphysics in this sense. Every question has its presuppositions. These are usually answers to logically prior questions; but if they are not, they are \textbf{absolute presuppositions.} Certain absolute presuppositions may be specific to a particular mode of inquiry. To the scientist engaging in that mode of inquiry, the presuppositions are unquestioned; however, a person is not simply a particular kind of scientist. One may step back as a philosopher, to identify the absolute presuppositions of oneself or others through \textbf{metaphysical analysis.} Some kind of unease may have driven one to engage in this analysis. The unease may be due to a conflict in one's presuppositions. In this case, when they are recognized, they will go away. The term ``metaphysical analysis'' is Collingwood's. The analogy with psychoanalysis is patent; however, Collingwood does not spell it out, though he underwent fifty sessions of analysis in 1938 \cite[p.\ 240]{Collingwood-Auto-2}. Being metaphysics in Collingwood's sense, our analysis of causation (as we follow along with Collingwood) will be historical. We use the word ``cause'' in at least three senses, which are historically related. Our concern is with the senses, not the word itself, in whose place we may use others such as ``force'' with similar senses. Getting the senses straight should improve our practice of science. In the original sense, it is \textbf{actions} that have causes. We perform an action---we \emph{act}---when we \begin{compactenum}[(1)] \item find ourselves in a situation, and \item intend to create a new one. \end{compactenum} These two factors compose the \textbf{cause in sense I} of our action. If the finding and intending are our own work, then we are the cause of our own action. Some other person may also be the cause, either by giving us to understand our situation, or persuading us to form an intention about it. If the situation that we want to create is, for example, an illuminated room at night, then we may learn to achieve this by throwing a switch on the wall. In sense I, \emph{we} are the cause of the illumination; but the \textbf{cause in sense II} is the switch. Metaphorically, at least, we transfer our agency to the switch. A cause in sense II is something in nature that we can use to bring about something else. A virus causes a disease in this sense, provided we can do something to prevent infection, or perhaps create it. If we can neither cure nor induce the disease, it has no cause in sense II. \emph{Practical} science seeks causes in sense II through \emph{experiment,} as distinct from mere observation. Once discovered, a cause in sense II will be expressed \emph{universally}: ``E always results from C.'' Different persons will recognize different causes of the same effect, depending on what those persons can severally accomplish regarding that effect. This is the principle of \textbf{relativity of causes.} One may however wish to rise above this relativity, assigning to each individual effect a unique cause. This then is a \textbf{cause in sense III.} The concept is ultimately incoherent, since there is no way to account for the necessity with which the cause is supposed to produce the effect. \begin{compactenum} \item This necessity is not mathematical, since we do not believe that conclusions about the world can be inferred without experiment. \item Neither is experiment enough: if something happens $n$ times, we still need a reason to conclude that it will happen again. \end{compactenum} The idea of necessitation---causation, compulsion---in the natural world is a remnant, a ``survival,'' of Neoplatonism, whereby God creates agents in the world to serve the divine purpose. People can have incoherent convictions. Kant and his followers are examples, for teaching both \begin{compactenum}[(1)] \item that every effect has its cause---% a unique cause, which therefore has sense III, and \item that this cause comes earlier in time---% and therefore has sense II. \end{compactenum} Natural science has learned to reject these teachings. Nonetheless, philosophers brought them into the twentieth century, thus threatening to inhibit the progress of science by creating hostility, both in the general public and in the academy. Thus would I summarize Collingwood's essay on causation in the \emph{Essay on Metaphysics.} According to the Author's Preface in the \emph{Essay,} the parts of Part III \begin{quote} might be called, as Descartes called a corresponding feature in a book of his own, \emph{specimina philosophandi.} One of them, that on causation, has already been printed in a different shape in the \emph{Proceedings} of the Aristotelian Society for the present year. \end{quote} The Preface is dated 2 April 1939. The \emph{Proceedings} volume that Collingwood refers to is \textsc{xxxviii,} of 1937--8. Collingwood's article there, ``On the So-Called Idea of Causation,'' was reprinted in the \emph{International Journal of Epidemiology,} 2014 \cite{Collingwood-IC}, along with two new articles of commentary \cite{Kronfeldner,Woodward}. It would be desirable to collate Collingwood's earlier paper with the essay in the \emph{Essay.} The latter is without a recommendation made in the former: \begin{quote} the best way of avoiding confusion will be to restrict our use of the word cause to occasions on which it is used in its ``proper'' sense, No.\ I; that on the occasions on which we use it in sense II we should be wise to use instead the terminology of means and ends; and that when we use it in sense III we should do better to speak of ``laws'' and their ``instances''. \end{quote} Collingwood himself ridicules this kind of recommendation in \emph{The Principles of Art} \cite[p.\ 255]{Collingwood-PA}: \begin{quote} philosophical controversies are not to be settled by a kind of police-regulation governing people's choice of words, and \lips a school of thought (to dignify it by that name) which depends for its existence on enforcing a particular jargon is a school which I neither respect nor fear. \end{quote} Again, Collingwood in the \emph{Essay} calls his work on causation a \emph{specimina philosophandi.} Today, in English, a specimen is an example or sample for examination; but a Latin dictionary translates \emph{specimen} as ``sign, evidence; token, symbol'' \cite{POLD}. This Latin sense is appropriate for a book of philosophy. As a sequence of words, a book is not itself philosophy, but the book may be a sign that philosophy has taken place. Similarly, by the account in \emph{The Principles of Art,} a painting is not a work of art, but a sign that art has taken place. It seems Descartes had \emph{Specimina Philosophiae,} ``Specimens of Philosophy''; Collingwood has replaced \emph{philosophiae} with \emph{philosophandi,} making the phrase mean something like ``specimens of how to do philosophy.'' My Latin is minimal; here I am just following examples like \emph{pudenda} ``that of which one ought to be ashamed,'' from \emph{pudeo} ``be ashamed.'' I do find Collingwood's work to be a specimen of how to do philosophy. Philosophy is a difficult pursuit, appropriately undertaken as a professional responsibility by certain academics. And yet philosophy is too important to be relegated to a university department. Collingwood makes both of those points in the \emph{Essay.} He begins the Prologue of his earlier \emph{Speculum Mentis} \cite[p.\ 17]{Collingwood-SM} with the assertion, \begin{quote} All thought exists for the sake of action. We try to understand ourselves and our world only in order that we may learn how to live. \end{quote} Collingwood effectively tries to live up to at least part of Thoreau's description in \emph{Walden} \cite[pp.\ 14--5]{Thoreau-Walden-Princeton}: \begin{quote} To be a philosopher is not merely to have subtle thoughts, nor even to found a school, but so to love wisdom as to live according to its dictates, a life of simplicity, independence, magnanimity, and trust. It is to solve some of the problems of life, not only theoretically, but practically. \end{quote} Collingwood can be irascible in a late work like the \emph{Essay.} He has started to suffer the strokes that will kill him in a few years, before he reaches the age of 54. He is set on edge by the coming war. He writes on page \pageref{fate}, ``The fate of European science and European civilization is at stake,'' as we try to get our metaphysics right. Others do not get metaphysics or Collingwood right. I am going to talk about two examples in the remainder of this Introduction. They may arise from resistance to the \emph{practicality} of Collingwood's thought. Simon Blackburn may be right about some persons' emotional response: \begin{quote} All this is off-putting,\label{off-putting} and Collingwood's readers have to learn to shake their heads with a smile rather than toss the whole thing in the bin. \end{quote} This is from a 2010 article \cite{Blackburn}, occasioned by the publication of a ``warm-hearted, affectionate biography of an irascible but brilliant philosopher and historian.'' The work is Fred Inglis's \emph{History Man} \cite{Inglis}, and Blackburn remarks on it, \begin{quote} Even if Collingwood was not the jovial, beer-drinking common man that Inglis would have liked him to be, it is good to see him brought some way back to the human fold. \end{quote} Having read the biography, I can agree with Blackburn's assessment. Blackburn also provides a good overview of Collingwood's work: \begin{quote} Most contemporary philosophers\lips %think of their subject in an entirely unhistorical way. They conceive of themselves as investigating things such as the nature of thought, or truth, or reason, or meaning. They wonder what a language, or a mind, or a world is that thought and reason and the rest are possible. This conception of the investigation is entirely ``a priori'': the problem would be the same as it perplexes us and as it perplexed Plato or Descartes. For Collingwood, this is all self-deception. What we may think of as a priori and timeless will be no such thing. It will be simply an application of the ``absolute presuppositions'' of our own period of thought. \end{quote} Concerning these absolute presuppositions, Blackburn himself engages in self-deception, saying that they \begin{quote} lie so far underneath the edifices we build that we cannot dig down to them. They remain invisible, if only because they would be at work determining the shape our digging would take, or what we could notice as we conducted it. We can never step on our own shadow. The only power that can reveal these presuppositions is that of time: later generations will see them, but we cannot. \end{quote} On the contrary, if we can discover the absolute presuppositions of past generations, then we can discover our own. Collingwood will make a similar suggestion about causation on page \pageref{own}: if somebody else can cause us to do something, we can also cause ourselves to do it. Learning our own absolute presuppositions is difficult, but possible. Collingwood is explicit on page 43 (in the original pagination) of the \emph{Essay}: \begin{quote} \lips in our less scientific moments, when knowledge appears to us in the guise of mere apprehension, intuiting that which simply confronts us, we are not even aware that whatever we state to ourselves or others is stated in answer to a question, still less that every such question rests on presuppositions, and least of all that among these presuppositions some are absolute presuppositions. In this kind of thinking, absolute presuppositions are certainly at work; but they are doing their work in darkness, the light of consciousness never falling on them. \emph{It is only by analysis that any one can ever come to know either that he is making any absolute presuppositions at all or what absolute presuppositions he is making.} \end{quote} I have italicized the last sentence. With too much appeal to formal logic, one might say that, by Collingwood's account, ``analysis'' is only a \emph{necessary} condition for coming to know one's absolute presuppositions; this does not mean coming to know them is actually possible. However, in an end-of-chapter note referring back to the passage above, Collingwood says, \begin{quote} People are not \emph{ordinarily} aware of their absolute presuppositions (p.\ 43), and are not, therefore, thus aware of changes in them; such a change, therefore, cannot be a matter of choice. \end{quote} Again the italics are mine, to highlight the exception that proves the rule: awareness of our own absolute presuppositions is \emph{possible,} through analysis, as we said. At least we can discover our absolute presuppositions as they were, just before we discovered them. Blackburn suggests that being discovered will change the presuppositions, by something like the Heisenberg Uncertainty Principle. Discovery \emph{may} effect a change in our presuppositions, owing to their heretofore unrecognized incompatibility. The conflict in them may disappear, like a neurosis under psychoanalysis. Collingwood describes such a conflict on the (original) page 96 of the \emph{Essay}: \begin{quote} The `science' [in the nineteenth century] which was to be protected by this cry of `No more Metaphysics' was being in effect described as a reactionary science, one which could only be imperilled by a critical inquiry into its foundations. Behind that cry there lay a feeling that the constellation of absolute presuppositions made by this reactionary science was exposed to certain strains\label{strain} which could only be `taken up' by keeping them in darkness. If people became aware that in certain contexts they were in the habit\label{habit} of treating this or that presupposition as an absolute one, they would be unable to go on doing it. \end{quote} For examples of such strains, I propose Oedipus and \emph{Marjory Morningstar} in note \ref{strain2}, page \pageref{strain2}. Collingwood's main example, to be considered in Chapter XXXIII (page \pageref{ch33}), is Kant's metaphysical analysis whereby \begin{inparaitem} \item[(\emph a)] every event has a cause, \item[(\emph b)] which is a previous event. \end{inparaitem} That this was generally accepted in the nineteenth century ``is strong, though of course not necessarily conclusive, evidence that it was correct'' (page \pageref{evidence})---% correct in the sense that persons really made these two presuppositions; and in this case, \emph{pace} Blackburn, they could see that they did. They made the presuppositions under strain though, because they were incompatible. Further analysis should reveal the incompatibility, thus freeing the former believers. Science is hampered by attempts to impose the presuppositions of a bygone era. Scientists can defend themselves by recognizing clearly what it is they are really doing. In this they can be aided by metaphysicians who understand their \emph{own} work. This is Collingwood's theme. Blackburn may acknowledge Collingwood's theme, while not thinking it of much importance. He says, \begin{quote} Collingwood hated the dominant philosophy of his time because of its unhistorical nature. It is not possible even to do science, in his view, without presupposing history, since it is only through their records and their results that scientists can pick up and profit from the labors of their colleagues. \emph{But in principle, at least, it is possible to be a good scientist, at the cutting edge of a field, with little historical sense.} \end{quote} In the last sentence (italicized by me), Blackburn uses ``history'' in the conventional sense, narrower than Collingwood's philosophical sense, described in general terms in \emph{An Essay on Philosophical Method} \cite[p.\ 35]{Collingwood-EPM}: \begin{quote} when a concept has a dual significance, philosophical and non-philosophical, in its non-philosophical phase it qualifies a limited part of reality, whereas in its philosophical it leaks or escapes out of these limits and invades the neighbouring regions, tending at last to colour our thought of reality as a whole. \end{quote} For each field of science, there may be some number $N$ such that what happened in the field $N$ years ago is of little use. However, what happened less than $N$ years ago is still history, and this is essential for the science. Blackburn seems only to acknowledge this when the science is \emph{philosophy,} which, \begin{quote} by contrast, is concerned with thought---% and as we have seen, Collingwood held that you cannot identify a thought unless you know to what question it was supposed to be an answer. The history of philosophy is therefore not a somewhat down-market curriculum option of no great interest to contemporary practitioners, just as the history of physics might exist alongside physics as something for retirement or bedtime. Instead \emph{Collingwood sees thought as something that is historically embodied}\lips We can understand where we are only by understanding where we have been\lips \end{quote} We can understand where we have been only through the traces it has left where we are now. \emph{This} is what makes history practical, by Collingwood's account in \emph{An Autobiography} \cite[p.\ 106]{An Autobiography}: \begin{quote} If the function of history was to inform people about the past, where the past was understood as a dead past, it could do very little towards helping them to act; but if its function was to inform them about the present, in so far as the past, its ostensible subject-% matter, was incapsulated in the present and consti\-% tuted a part of it not at once obvious to the untrained eye, then history stood in the closest possible relation to practical life. \end{quote} Some of these traces may be \emph{survivals,} as Collingwood describes them on page \pageref{survivals} below and, at greater length, in a manuscript he did not publish himself \cite[p.\ 142]{Collingwood-PM}: \begin{quote} Civilized peoples have developed out of savage ones; and civilization contains many elements which, taken at their face value, are condemned as irrational and described as superstitions; but it would be better to drop that word and describe them as survivals, for they are things whose proper home and meaning must be sought in the context of an earlier civilization. \end{quote} Though I am a fan both of Collingwood and of my \emph{alma mater} St John's College \cite{Pierce-SJC}, the latter may \emph{not} treat thought ``as something that is historically embodied,'' in the clause of Blackburn that I italicized. At the College, there is no study of history as such, although there is study of mathematics, language, science, and music. And yet the whole Program is historical, if only in the sense that its texts are read mostly in chronological order. There is no ``literature segment,'' in which one reads together Homer, Dante, and Jane Austen; no ``metaphysics segment,'' in which one reads Aristotle, Descartes, and Kant. One reads them all, but over a span of years, in the order: Homer, Aristotle, Dante, Descartes, Kant, Jane Austen. By contrast, in a conventional philosophy department, one may take a course in metaphysics. Then one might read a textbook such as Michael J. Loux, \emph{Metaphysics: A contemporary introduction} \cite{Loux}, which I happened upon in an Ankara bookshop some years ago. Collingwood begins his own \emph{Essay on Metaphysics} by analyzing Aristotle's book on the subject. He writes down (on his page 11) two propositions, ``each of which offers what might be called a definition of metaphysics'': \begin{quote} \begin{compactenum} \item Metaphysics is the science of pure being. \item Metaphysics is the science which deals with the presuppositions underlying ordinary science; where by `ordinary science' I mean such thinking as is `scientific' in the sense defined in the preceding chapter, and `ordinary' in the sense that is it not a constituent part of metaphysics. \end{compactenum} \quad In this chapter I shall argue that the first of these two propositions cannot be true because a science of pure being is a contradiction in terms. The second proposition I take to be true, and this book as a whole represents my endeavour to explain its meaning. \end{quote} Loux makes the same distinction, but his choice of focus is opposite to Collingwood's. In Kant's terminology, the two understandings of metaphysics are ``transcendent'' and ``critical,'' and by Loux's account, \begin{quotation} \noindent Whereas transcendent metaphysics seeks to characterize a reality that transcends sense experience, critical metaphysics has as its task the delineation of the most general features of our thought and knowledge\lips Kant's conception of a metaphysical enterprise whose task it is to identify and characterize the most general features of our thought and experience is one that continues to find defenders in our own day. \end{quotation} Now Loux makes a note, providing evidence for Blackburn's speculation about Collingwood: ``I doubt if he is more than a ghost in the footnotes to syllabi across the Western world.'' Loux's note is not even a footnote, but an end-of-chapter note. There he makes the only reference to Collingwood in his book---% at least according to the Index. Neither the Index nor the note itself gives the page where the reference to the note is found. That page is 8; the note itself is 8 among the nine notes on page 17. In his note, Loux lists Collingwood's \emph{Essay} as the earliest of several ``examples of this [critical] approach to metaphysics.'' Then he continues in the main text: \begin{quote} These philosophers tell us that metaphysics is a descriptive enterprise whose aim is the characterization of our \emph{conceptual scheme} or \emph{conceptual framework.} \end{quote} Presently Loux assigns to these philosophers a jocular if not pejorative and dismissive title: ``conceptual schemers.'' Loux will not follow them: \begin{quote} I am inclined to think that traditional metaphysicians are right here. As I see it, arguments designed to undermine the conception of metaphysics as traditionally understood invariably call themselves into question. In any case, it is metaphysics as traditionally understood that we will be doing or trying to do in this book. The aim will be to characterize the nature of reality, to say how things are. \end{quote} I would say rather that \emph{skepticism}\label{scepticism} calls itself into question, as Collingwood points out in \emph{An Essay on Philosophical Method} \cite[pp.\ 140--1]{Collingwood-EPM}: \begin{quote} Scepticism\lips is in reality a covert dogmatism; it contains positive theories of the nature, method, and limitations of philosophical thought, but disclaims their posses\-% sion and conceals them from criticism. Hence it is both inconsistent, or false to its own professed principles, and---intentionally or unintentionally---% dishonest, because applying to others a form of criticism which in its own case it will not admit. \end{quote} (See also note \ref{scepticism2}, page \pageref{scepticism2}.) Meanwhile, Loux has said: \begin{quote} If the conceptual schemer is correct in claiming that the act of conceptual representation bars us from an apprehension of anything we seek to represent, then why should we take seriously the schemer's claims about the conceptual representation? \end{quote} Why indeed? But the kind of skeptical stance that Loux attributes to his ``conceptual schemers'' is not Collingwood's. Again Loux: \begin{quote} Traditional metaphysicians will go on to insist that we manage to think and talk about things---% things as they really are and not just things as they figure in the stories we tell. They will insist that the very idea of thinking about or referring to things presupposes that there are relations that tie our thoughts and words to the mind-independent, language-independent things we think and talk about\lips \end{quote} If ``traditional metaphysicians'' are insisting that we have certain presuppositions, then these persons may indeed be metaphysicians, by Collingwood's account, as summed up in the last quotation above from the \emph{Essay on Metaphysics,} as well as on its original page 47: \begin{quote} Metaphysics is the attempt to find out what absolute presuppositions have been made by this or that person or group of persons, on this or that occasion or group of occasions, in the course of this or that piece of thinking. \end{quote} But if Loux's ``traditional metaphysicians'' assert that their own presuppositions are everybody's, throughout time, well, evidently this is a presupposition too, and it ought to be brought to light. Loux seems to have missed the point of Collingwood, noted by Blackburn: the project of saying how things are is certainly \emph{possible}: but it is \emph{historical.} The point should be developed, but I am not going to try to do it here. I shall just note how I think Loux is at best misguided. He says in the first paragraph of his Chapter 1 (called ``The problem of universals I---Metaphysical realism''): \begin{quote} Although almost everyone will concede that some of our ways of classifying objects reflect our interests, goals, and values, few will deny that many of our ways of sorting things are fixed by the objects themselves.%%%%% \footnoteA{An exception, of course, is the conceptual schemer we discussed in the Introduction. [Loux's note.]} %%%%%%%%%%%%%%%%%%%% It is not as if we just arbitrarily choose to call some things triangular, others circular, and still others square; they \emph{are} triangular, circular, and square. Likewise, it is not a mere consequence of human thought or language that there are elephants, oak trees, and paramecia. They come that way, and our language and thought reflect these antecedently given facts about them. \end{quote} Collingwood himself considers the notion of arbitrary choice, beginning on page \pageref{arbitrary} below. Obviously our manner of talking about things is not arbitrary. But to me it is obvious as well that things don't just ``come'' the way we talk about them. Consider: ``People come in two races, black and white: this is no mere consequence of human thought or language.'' I am sure that there are many people who would agree with this: but they are mistaken. The way we classify things may be based on the things, but they do not tell us how to do it. \mychap[Note on the Text]{Note on the Text} Any version of an old text should provide an explanation of how the text was obtained and edited, so that readers may assess its authenticity. I have taken Collingwood's text from a \url{pdf} scan, found on the Web, of a reissuing \cite{Collingwood-EM1} of the first edition. The colophon there includes: \begin{center} \smaller \textsc{first edition 1940}\\\vspace{0.5\baselineskip} Reprinted photographically in Great Britain\\ at the Oxford University Press, 1948\\ from sheets of the first edition \end{center} In particular, there is no assertion of copyright. Presumably this is because of Collingwood's express opposition to copyright in \emph{The Principles of Art} (1938). However, Collingwood died in 1943. I used an optical character recognition (\textsc{ocr}) program to convert the desired pages of the \url{pdf} file of the \emph{Essay} into a text file. I then made the text file into a \LaTeX\ file. Editing this involved doing the following. \begin{compactitem} \item Removing the scanned page headings.%%%%% \footnoteC{In the original, the heading of each page that does not begin a chapter consists of the name of the chapter, in Chapters XXIX, XXX, and XXXIV. In chapters XXXI, XXXII, and XXXIII, the name being too long for one page, it is divided across each two-page spread: \begin{quote} \begin{tabular}{cc} CAUSATION IN PRACTICAL&NATURAL SCIENCE\\ CAUSATION IN THEORETICAL&NATURAL SCIENCE\\ CAUSATION IN&KANTIAN PHILOSOPHY \end{tabular} \end{quote} The last page of chapter XXXI being even, it has the heading ``PRACTICAL NATURAL SCIENCE.'' My pages being wider (typographically speaking) than Collingwood's, I need not divide chapter titles. But a \LaTeX\ package might be desirable that provided a command with five arguments: \begin{inparaenum}[(1)] \item chapter title; \item first half of title, for heading even pages inside the chapter; \item second half of title, for odd pages inside the chapter; \item abbreviated title, for the last page of the chapter, if even; \item abbreviated title for the table of contents. \end{inparaenum} The existing \url{\chapter} command takes only two arguments: title and abbreviated title.} %%%%%%%%%%%%%%%%%%%%%%% \item Marking up footnotes and \emph{italics} as such. \item Replacing the ligatures ``fl'' and ``fi'' (often ill-scanned) with distinct letters \url{fl} and \url{fi} (which the \TeX\ program then makes into ligatures again; apparently the \textsc{ocr} program recognized no ``ff'' or ``ffi'' ligatures). \item Resetting the first word of each chapter in \textsc{small capitals,} as in the original. \item correcting the instances of Greek text (which the \textsc{ocr} program did not recognize at all). \item Following abbreviations like \texttt{e.g.}, \texttt{i.e.}, \texttt{loc.}, \texttt{Mr.}, \texttt{op.}, and \texttt{pp.}\ with \texttt{\textbackslash\textvisiblespace} (backslash followed by space) so that \TeX\ knows that they do not end a sentence. \item Removing, or replacing with a hyphen, the text file's \texttt{\textasciitilde\textvisiblespace} (tilde followed by space, evidently used by the \textsc{ocr} program in place of the hyphen at the end of a line). \item Making other corrections, such as when the scanner confuses ell with one (l with 1) or oh with zero (o with 0). \end{compactitem} \addpart[R. G. Collingwood, ``Causation'']% {R. G. Collingwood\\``Causation''\\from \emph{An Essay on Metaphysics}} \refoot{\emph{Essay on Metaphysics}, Part III\textsc c (Causation)} %\setcounter{chapter}{28} %\renewcommand{\thechapter}{\Roman{chapter}} \mychap[XXIX. Three Senses of the Word `Cause']{XXIX\\ THREE SENSES OF THE WORD `CAUSE'} \textsc{Conformably} to the historical nature of metaphysics, any discussion of a metaphysical difficulty must be historically conducted. One major difficulty, or group of difficulties, now exercising students of metaphysics is connected with the idea of causation. I do not hope in the present part of my essay to offer a complete solution for this difficulty or group of difficulties; all I propose to do is to show what I mean by saying that it ought to be discussed historically. I shall confine myself to making two main points. 1. That the term `cause',% \label{123} as actually used in modern English and other languages, is ambiguous. It has three senses; possibly more; but at any rate three.%%%%% \footnoteB{\label{OED-def}Collingwood's three senses correspond roughly, in reverse order, to definitions 1--3 of ``cause'' in the \emph{Oxford English Dictionary} \cite{OED}. For definition 2, see note \ref{OED2}, page \pageref{OED2}; definition 1, note \ref{OED1}, page \pageref{OED1}; definition 3, note \ref{OED3}, page \pageref{OED3}. The remaining \emph{OED} definitions, 4--12, are for obsolete or legal senses, or for ``cause'' as used in set phrases. Evidently Collingwood has consulted the \emph{OED}: see page \pageref{OED}.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Sense I. Here that which is `caused' is the free and deliberate act of a conscious and responsible agent, and `causing' him to do it means affording him a motive for doing it.%%%%% \footnoteB{\label{Greenwald}%%% Collingwood will provide his own examples when he takes up the three senses of ``cause'' respectively in the next three chapters. Meanwhile, I suggest that Glenn Greenwald uses sense I of ``cause'' in an article by called ``Canada, at war for 13 years, shocked that a `terrorist' attacked its soldiers'' \cite{Greenwald-Canada}: \begin{quote} The issue here is not justification (very few people would view attacks on soldiers in a shopping mall parking lot to be justified). The issue is causation\lips Except in the rarest of cases, the violence has clearly identifiable and easy-to-understand causes: namely, anger over the violence that the country's government has spent years directing at others. \end{quote} Terrorist violence is an example of what Collingwood calls a `` free and deliberate act of a conscious and responsible agent''; nonetheless, it will have motives, and these would seem to be what Greenwald refers to as ``causes.'' Some persons would seem to be resistant to this terminology, as Greenwald points out in an Update to the original article: \begin{quote} no matter how clear you make it that you are writing about causation and not justification, many will distort what you write to claim you've justified the attack. \end{quote} Such disputes as between Greenwald and his detractors are one cause of the study the text of Collingwood presented here. Greenwald himself uses ``cause'' in Sense II in the same Update; see the next note, along with note \ref{Greenwald2} on page \pageref{Greenwald2}.} %%%%%%%%%%%%%%%%%%%%%%%%%%%% Sense II. Here that which is `caused' is an event in nature, and its `cause' is an event or state of things by producing or preventing which we can produce or prevent that whose cause it is said to be.%%%%% \footnoteB{In the Update of the article mentioned in the previous note, Greenwald writes also, \begin{quote} If one observes that someone who smokes four packs of cigarettes a day can expect to develop emphysema, that's an observation about causation, not a celebration of the person's illness. \end{quote} I take the emphysema to be an event in nature, and its cause is smoking. However, if we mean cause in Sense II, Greenwald's assertion here too will be controversial, if some persons wish to deny that smokers have any control over their habit.} %%%%%%%%%%%%%%%%%%%%%%% Sense III. Here that which is `caused' is an event or state of things, and its `cause' is another event or state of things standing to it in a one-one%%%%% \label{one-one} %%%%%%%%%%%%%% relation of causal priority: i.e.\ a relation of such a kind that (\emph a) if the cause happens or exists the effect also must [\textbf{286}] happen or exist, even if no further conditions are fulfilled, (\emph b) the effect cannot happen or exist unless the cause happens or exists, (\emph c) in some sense which remains to be defined, the cause is prior to the effect; for without such priority there would be no telling which is which. If C and E were connected merely by a one-one relation such as is described in the sentences (\emph a) and (\emph b) above, there would be no reason why C should be called the cause of E, and E the effect of C, rather than vice versa. But whether causal priority is temporal priority, or a special case of temporal priority, or priority of some other kind, is another question.%%%%% %%%%% \footnoteB{\label{implies}% It is not clear at this point why the same effect cannot have any one of a number of causes, even in sense III where there is no implication of human involvement. Tears can be caused by sadness or onions; dark clouds, by burning tires or by dust storm. It might be said then that we are talking not about the same effect, but the same \emph{kind} of effect. Still, this would seem to be allowed: as Collingwood will say on page \pageref{indiv-gen},\label{indiv-gen-cite} ``In sense III causal propositions might equally well be either individual or general.'' Meanwhile, If C \emph{implies} E logically, then each of C and E is called a \emph{condition} of the other, but there is still a distinction: C is a \emph{sufficient} condition for E, and E is a \emph{necessary} condition for C. Collingwood will take up the relation between causation and implication in Chapter XXXII, starting on page \pageref{implication}.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \uline{Sense I may be called the \emph{historical} sense of the word `cause'}, because it refers to a type of case in which both C and E are human activities such as form the subject-matter of history. When historians talk about causes, this is the sense in which they are using the word, unless they are aping the methods and vocabulary of natural science.%%%%%%%%%%%%%% \footnoteB{Collingwood uses the verb ``ape'' here pejoratively. Historians ought not to confuse their work with natural science. See for example \emph{The Principles of History} \cite[p.\ 83]{Collingwood-PH}: \begin{quote} \lips all the human sciences distinguish what a thing is meant to be from what it is, and aim at distinguishing cases where the two coincide (successes) from cases in which they do not (failures). But according to the assumptions of our natural science\lips cases of this kind are not distinguishable in the world of nature\lips \end{quote}} %%%%%%%%%%%%%%%%%%%%%%%%%%%% \uline{Sense II refers to a type of case in which natural events are considered from a human point of view}, as events grouped in pairs where one member in each pair, C, is immediately under human control, whereas the other, E, is not immediately under human control but can be indirectly controlled by man because of the relation in which it stands to C. This is the sense which the word `cause' has in the \emph{practical sciences of nature}, i.e.\ the sciences of nature whose primary aim is not to achieve theoretical knowledge about nature but to enable man to enlarge his control [\textbf{287}] of nature. This is the sense in which the word `cause' is used, for example, in engineering or medicine.%%%%% \footnoteB{\label{OED2}% The closest the \emph{OED} comes to this sense of cause seems to be in the second definition: \begin{quotation} \textbf{2.} A person or other agent who brings about or occasions something, with or without intention. (Often in bad sense: One who occasions, or is to blame for mischief, misfortune, etc.) \emph c \textbf{1374} \textsc{Chaucer} \emph{Anel.\ \&\ Arc.}\ [\emph{Anelida and Arcite}] 257 \TH aughe \th at yee \TH us Causelesse \th e Cause be Of my dedely aduersitee\lips \end{quotation}} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \uline{Sense III refers to a type of case in which an attempt is made to consider natural events} not practically, as things to be produced or prevented by human agency, but \uline{theoretically}, as things that happen independently of human will but not independently of each other: causation being the name by which this dependence is designated. This is the sense which the word has traditionally borne in physics and chemistry and, in general, the \emph{theoretical sciences of nature}.%%%%% \footnoteB{\label{OED1}% This is roughly the first definition in the \emph{OED}: \begin{quotation} \textbf{1.} That which produces an effect; that which gives rise to any action, phenomenon, or condition. \emph{Cause} and \emph{effect} are correlative terms. \emph c \textbf{1315} \textsc{Shoreham} 117 Cause of alle thyse dignyte . . Was Godes owene grace\lips \textbf{b.} as philosophically defined. \textbf{1656} tr.\ \emph{Hobbes' Elem.\ Philos.}\ \textsc{ii} ix.\ (1839) 121 A cause simply, or an entire cause, is the aggregate of all the accidents both of the agents how many soever they be, and of the patient, put together; which when they are all supposed to be present, it cannot be understood but that the effect is produced in the same instant \lips \end{quotation}} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% The difficulties to which I referred at the beginning of this chapter are all connected with sense III. The other two senses are relatively straightforward and easy to understand. They give rise to no perplexities. The only perplexities that ever occur in connexion with them are such as arise from a confusion of sense I with sense II, or from a confusion of either with sense III.%%%%%% \footnoteB{\label{Greenwald2}%%% Collingwood seems wrong here, because there are such controversies as discussed in note \ref{Greenwald}, page \pageref{Greenwald}. When the citizenry are accused of having brought on the terrorist attack that they have suffered, they object, perhaps because they think responsibility is being \emph{transferred} from the attacker to themselves. However, sense I of causation implies a \emph{sharing} of responsibility, according to Collingwood on page \pageref{share}. It might mollify the sufferers of the attack to understood the original accusation as involving sense II of causation. This would mean the attack on them was ``an event in nature,'' which they might have worked to prevent, as they might work to prevent or ameliorate the destructive effects of an earthquake. Still, in the immediate aftermath of a disaster, nobody wants to be told what they \emph{should have} done.} %%%%%%%%%%%%%%%% But sense III, as I shall show, raises difficult problems quite by itself, and apart from any confusion with other senses. These problems are due to internal conflict.%%%%% \label{internal-conflict}%%%%% \footnoteB{\label{RP}% Collingwood took up these problems, even in his first book, \emph{Religion and Philosophy,} particularly in Part II, ``Religion and Metaphysics''; Chapter II, ``Matter''; \S2, ``Materialism''; (b), ``The paradox of causation'' \cite[pp.\ 82--9]{Collingwood-RP}. See the passages about \begin{inparaenum}[(1)] \item the universe as self-causing (note \ref{RP-whole}, page \pageref{RP-whole}); \item everything's being willed (note \ref{RP-Granted}, page \pageref{RP-Granted}); \item the dualism of ``popular metaphysic'' (note \ref{RP-Popular}, page \pageref{RP-Popular}); \item a tree's fall (note \ref{RP-tree}, page \pageref{RP-tree}). \end{inparaenum}} %%%%%%%%%%%%%%%%%%%%%%%%%%% The various elements which go to make up the definition of sense III are mutually incompatible. This incompatibility, at the lowest estimate, constitutes what I called in Chapter VII a `strain' in the current modern idea of causation, and therefore in the whole structure of modern natural science in so far as modern natural science is based on that idea.%%%%% \footnoteB{This then is the main theme of Collingwood's present text: the history of modern science. Freedom and responsibility compose a theme of Collingwood's \emph{New Leviathan} \cite{Collingwood-NL}, mentioned in notes \ref{NL-law} (page \pageref{NL-law}) and \ref{NL-interact} (page \pageref{NL-interact}).} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% I have called I, II, and III different `senses' of the [\textbf{288}] word `cause'. A technical objection might be lodged against this expression on any of three grounds, if no more. (\emph a) `What you have distinguished are not three senses of the word ``cause'', but three types of case to any one of which that word is appropriate, the sense in which it is used being constant.' But, as I shall try to show, if you ask what exactly you mean by the word on each type of occasion you will get three different answers. (\emph b) `What you have distinguished is three kinds of causation.' But the three definitions of causation referred to in the foregoing paragraph are not related to each other as species of any common genus; nor is there any fourth definition, the definition of cause in general, of which the three `kinds' of causation are species.%%%%% \footnoteB{This is an echo of the ``overlap of classes''\label{overlap} discussed in Collingwood's \emph{Essay on Philosophical Method} \cite{Collingwood-EPM}. When Collingwood makes a sharp distinction between the senses of cause, he ought to have a good reason. He has already said that there is something fuzzy about sense III.} %%%%%%%%%%%%% (\emph c) `One of your three so-called senses of the word ``cause'' is the only proper sense; the other two represent metaphorical usages of the word.' In order to show how baseless this objection is, it would be necessary to show that the distinction between `proper' and `metaphorical' senses of words is illusory. The contradictory of `proper' is not `metaphorical' but `improper'. A proper%%% \label{proper} usage of a word is one which as a matter of historical fact occurs in the language to which the word belongs. The contradictory of `metaphorical' is `literal'; and if%%%%%%% \label{metaphor} %%%%%%%%%% the distinction between literal and metaphorical usages is a genuine distinction, which in one sense it is, both kinds of usage are equally proper. There is another sense in which all [\textbf{289}] language is metaphorical; and in that sense the objection to certain linguistic usages on the ground that they are metaphorical is an objection to language as such, and proceeds from an aspiration towards what Charles Lamb called the uncommunicating muteness of fishes. But this topic belongs to the theory of language, that is, to the science of aesthetic, with which this essay is not concerned.%%%%% \footnoteA{The main questions involved, as I see them, are discussed in my \emph{Principles of Art,} especially Chapter XI.}%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%% \footnoteB{In the Editor's Introduction to the Revised Edition of the \emph{Essay} \cite[p.\ xci, n.\ 23]{Collingwood-EM}, Rex Martin suggests that Collingwood did take senses II and III of ``cause'' to be metaphorical, in his original paper. There he says, for example \cite[p.\ 1697]{Collingwood-IC}, \begin{quote} Sense I, which is historically the original sense, is presupposed by the others, and remains strictly speaking the one and only ``proper'' sense. When we assert propositions containing the word cause in senses II and III, we are ``saying'' one thing and ``meaning'' another\lips This always has an element of danger in it\lips This danger is much worse when our ``metaphors'' get ``mixed''. \end{quote} Though the original paper is dated 1937--8, Collingwood may have written it before feeling the need to rewrite \emph{Outlines of a Philosophy of Art.} He needed to rewrite, because he had ``changed his mind on some things,'' according to the Preface, dated 22 September 1937, of \emph{The Principles of Art.} Martin's note is to his paragraph \cite[p.\ lx]{Collingwood-EM}, \begin{quote} So we will examine initially only senses I and II, for they are the basic ones in Collingwood's account. Collingwood emphasizes here that these are two distinct senses of `cause'. But neither one is \emph{the} proper sense, in his view. The term `cause' is equivocal and these are simply two of its different meanings. \end{quote} This sounds as if Martin \emph{accepts} objection (\emph b), ``$\;$`What you have distinguished is three kinds of causation.'$\;$'' The word ``cause'' is not \emph{simply} equivocal, but its meanings are historically related. Moreover, though Martin says ``we will examine \emph{initially} only senses I and II, in an account of 14 pages (liv--lxvii), he hardly mentions sense III again, even though, as Collingwood said above, \begin{quote} The difficulties to which I referred at the beginning of this chapter are all connected with sense III. The other two senses are relatively straightforward and easy to understand. \end{quote} Martin skips sense III because \begin{inparaenum}[(1)] \item ``Collingwood's account of it is obscure and vaguely worded,'' \item the sense is ``somehow derivative from the other two,'' and \item it is ``no longer figuring significantly in the most advanced theoretical science, physics'' \end{inparaenum} \cite[p.\ lx]{Collingwood-EM}.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% At the same time I do not wish to imply that the distinction between I, II, and III is an example of what Aristotle calls `accidental equivocation'.%%%%% \footnoteA{\emph{Eth.\ Nic.}\ 1096\textsuperscript b 26--7: (although the various goodnesses of honour, wisdom, and pleasure are not identical in definition but differ \emph{qua} goodnesses) `the case does not resemble one of accidental equivocation', \gr{o>uk >'eoike to~is ge >ap`o t'uqhs it'ia}, we find that sense I is not only an established modern sense, it is also of great antiquity. In English it goes back, as the quotations in the \emph{Oxford English Dictionary}\label{OED} show, to the Middle Ages.%%% \footnoteB{\label{OED3}The third definition of ``cause'' in the \emph{OED} is, \begin{quote} A fact, condition of matters, or consideration, moving a person to action; ground of action; reason for action, motive. \end{quote} The first illustrative quotation is from the \emph{Ancren Riwle,} dated before 1225. See Appendix \ref{app:AR}, page \pageref{app:AR}.} %%%%%%%%%%%%%%%%%% In Latin it is the commonest of all the senses distinguished by Lewis and Short, and also the oldest. In Greek, as the articles \gr{a>it'ia, a>'itios} in Liddell and Scott show, the word which in Latin is translated \emph{causa} meant originally `guilt', `blame', or `accusation', and when first it began to mean `cause', which it sometimes does in fifth-century literature, it was used in sense I, for the cause of a war or the like. In fact, the historical relation between these senses is the opposite of what has been suggested. Sense I is the original sense, and senses II and III have been derived from it by a process I shall trace in the sequel. [\textbf{292}] 2. Sense I does not logically presuppose sense III. On the contrary, as I shall show in the following chapters, both sense II and sense III logically presuppose sense I; and \uline{any attempt to use the word in sense II or III without the anthropomorphic implications belonging to sense I must result either in a misuse of the word cause} (that is, its use in a sense not consistent with the facts of established usage), \uline{or in a redefinition of it} so as to make it mean what in established usage it does not mean: two alternatives which differ only in that established usage is defied with or without a formal declaration of war.%%%%% \footnoteB{This is the core idea. The misuse or redefinition of ``cause'' is not simply a linguistic problem, of concern only to a pedant such as Collingwood might be accused of being; but it leads to the ``internal conflict'' mentioned on page \pageref{internal-conflict}.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \uline{A cause in sense I is made up of two elements, a \emph{causa quod}%% \label{quod} or efficient cause and a \emph{causa ut} or final cause}.%%%%% \footnoteB{\label{4causes}%%%%%%%%%% See Appendix \ref{app:Aristo} on Aristotle's causes.} %%%%%%%%%%%%%%%% The \emph{causa quod} is a situation or state of things existing; the \emph{causa ut} is a purpose or state of things to be brought about. Neither of these could be a cause if the other were absent. A man who tells his stockbroker to sell a certain holding may be caused to act thus by a rumour about the financial position of that company; but this rumour would not cause him to sell out unless he wanted to avoid being involved in the affairs of an unsound business. And \emph{per contra} a man's desire to avoid being involved in the affairs of an unsound business would not cause him to sell his shares in a certain company unless he knew or believed that it was unsound. \uline{The \emph{causa quod}}\label{quod2} is not a mere situation or state of things, it \uline{is a situation or state of things known or believed by the agent in question to exist}.%%%%% \footnoteB{This is a theme of Collingwood's philosophy of history. Unlike natural science, history is not about ``situations or states of things'' as such, but about what is thought of them \cite[p.\ 93]{Collingwood-PH}: \begin{quote} It is not nature as such and in itself (where nature means the natural environment) that turns man's energies here in one direction, there in another: it is what man makes of nature by his enterprise, his dexterity, his ingenuity, or his lack of these things. \end{quote}} %%%%%%%%%% If a prospective litigant briefs a certain barrister because of [\textbf{293}] his exceptional ability, the \emph{causa quod} of his doing so is not this ability simply as such, it is this ability as something known to the litigant or believed in by him. \uline{The \emph{causa ut} is not a mere desire or wish, it is an intention}. The \emph{causa ut} of a man's acting in a certain way is not his wanting to act in that way, but his meaning to act in that way. There may be cases where mere desire leads to action without the intermediate phase of intention; but such action is not deliberate, and therefore has no cause in sense I of the word.% \footnoteB{\label{advertising}% \emph{Advertising} may be intended to create desire that leads to the action of spending, without deliberation on the part of the consumer. Non-deliberate action would seem to be ``natural,'' making advertising a cause in sense II. The ``natural'' effect of advertising would be the enrichment of the advertiser. The ensuing discussion suggests that ``non-deliberate action'' is a contradiction in terms, or at least does not refer to action in the fullest sense. See also note \ref{startling}, page \pageref{startling}.} %%%%%%%%%%%%%%%%%%%%%%%%%%% \uline{Causes in sense I of the word may come into operation%%% \label{come-into-op} through the act of a second conscious and responsible agent}, in so far as he (1) either puts the first in a certain situation in such a way that the first now believes or knows himself to be in that situation, or alternatively informs or persuades the first that he is in a certain situation; or (2) persuades the first to form a certain intention. In either of these two cases, the second agent is said to cause the first to do a certain act, or to `make him do it'. The act so caused is still an act; it could not be done (and therefore could not be caused) unless the agent did it of his own free will. If A causes B to do an act \gr b, \gr b is B's act and not A's; B is a free agent in doing it, and is responsible for it. If \gr b is a murder, which A persuaded B to commit by pointing out certain facts or urging certain expediencies, B is the murderer. There is no contradiction between the proposition that the act \gr b was caused by A, and [\textbf{294}] the proposition that B was a free agent in respect of \gr b, and is thus responsible for it. On the contrary, the first proposition implies the second.%%%%% \footnoteB{%\label{AB} If \gr b is a murder committed by B, then to say that A caused \gr b is to accept the proposition that B actually performed the act. In another situation, as when A secretly drugs B into unconsciousness and then wraps the hand of B around the pistol, then there is no act \gr b, but rather the act \gr a by A of performing the murder.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Nevertheless, in this case A is said to `share%%% \label{share} the responsibility' for the act \gr b.%%%%% \footnoteB{See note \ref{Greenwald2}, page \pageref{Greenwald2}, about confusion over the sharing of responsibility.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% This does not imply that a responsibility is a divisible thing, which would be absurd;%%%%%% \footnoteB{The \emph{corporation} divides a limited responsibility among shareholders.} %%%%%%%%%%%%%%%% it means that, whereas B is responsible for the act \gr b, A is responsible for his own act, \gr a, viz.\ the act of pointing out certain facts to B or urging upon him certain expediencies, whereby he induces him to commit the act \gr b. When a child accused of a misdeed rounds on its accuser, saying, `You made me do it', he is not excusing himself, he is implicating his accuser as an accessory. This is what Adam was doing when he said, `The woman whom thou gavest to be with me, she gave me of the tree, and I did eat'.%%%%% \footnoteB{\label{NL-law}A study of \emph{law} might be in order. Collingwood's \emph{New Leviathan} \cite{Collingwood-NL} is full of references to Roman law, such as the following, from Chapter XIX, ``Two Senses of the Word `Society.'$\;$'': \begin{quotation} \textbf{19.\ 5.} The word `society' in modern European languages is borrowed from the vocabulary of Roman law. \textbf{19.\ 51.} \emph{Societas} is a relation between \emph{personae} (that is, human beings capable of sueing and being sued, who must be free man and not slaves, Roman citizens and not foreigners, male and adult, not in the \emph{manus} or \emph{patria potestas} of another but heads of families) whereby they join together of their own free will in joint action. \end{quotation} Of the legal senses of ``cause'' in the \emph{OED} (note \ref{OED-def}, page \pageref{OED-def}), the basic one is, ``the matter about which a person goes to law.'' But again, our purpose is not to study the word ``cause'' as such (note \ref{tech}, page \pageref{tech}).} %%%%%%%%%%%%%%%%%%%%%% \uline{A man}%%%%% \label{own} %%%%%%% is \uline{said to act `on his own responsibility' or `on his sole responsibility'} when (1) his knowledge or belief about the situation is not dependent on information or persuasion from any one else, and (2) his intentions or purposes are similarly independent. In this case (the case in which a man is ordinarily said to exhibit `initiative') his action is not uncaused. It still has both a \emph{causa quod} and a \emph{causa ut}. But because he \uline{has done for himself, unaided, the double work of envisaging the situation and forming the intention}, which in the alternative case another man (who is therefore said to cause his action) has done for him, he \uline{can now be said to cause his own action as well as to do it}. If he invariably acted in that way the total complex of his activities could [\textbf{295}] be called self-causing%%%%% \label{self-cause} %%%%% (\emph{causa sui}); an expression which refers to absence of persuasion or inducement on the part of another, and is hence quite intelligible and significant, although it has been denounced as nonsensical by people who have not taken the trouble to consider what the word `cause' means.%%%%% \footnoteB{\label{RP-whole}\label{scepticism2}% Collingwood often declines to name the persons he disagrees with. This can be understood as politeness. What he is criticizing is the ``persona'' or mask of the author of some work, and not the \emph{person.} The thought of the person may have changed and developed since the writing. The person is not now present to offer a defense, a clarification, or a retraction. Nonetheless, Collingwood is going to disagree with certain named persons in Chapter XXXIII (page \pageref{ch33}). The idea of self-causing activity is seen in \emph{Religion and Philosophy} \cite[p.\ 87]{Collingwood-RP}, in (iii.)\ of the subsection mentioned in note \ref{RP} (page \pageref{RP}): \begin{quote} The first law of matter is that it cannot originate states in itself. But the universe as a whole, if it has any states, must originate them itself; and yet if it does so it breaks the first law of matter; for it is itself a material thing. But the universe only means all that exists; so if the universe is an exception to the law of causation, everything is an exception to it, and it never holds good at all. \end{quote} \emph{Matter} cannot originate states in itself; but a person can. One might argue on the contrary that, since everything \emph{is} matter, a person in particular is matter, and thus one \emph{cannot} originate states in oneself: any appearance otherwise is an illusion. This is the kind of self-defeating scepticism that I mention on page \pageref{scepticism}. We may \emph{presuppose} that all is matter; but we \emph{observe} is that we originate states in ourselves. We do things (see note \ref{strain2}, page \pageref{strain2}). An explanation of this observation that does away with it is no explanation.} %[\textbf{296}] \mychap[XXXI. Causation in Practical Natural Science]{XXXI\\ CAUSATION IN PRACTICAL NATURAL SCIENCE}%XXXI \textsc{In} sense I of the word `cause' that which is caused is a human action (including under that name actions of other, non-human, agents, if there are any, which act in the same conscious, deliberate, and responsible way which is supposed to be characteristic of human beings).%%%%% \footnoteB{\label{strain2}A cat may cause us, in sense I, to open a door for her. However, even among humans, action in a ``conscious, deliberate, and responsible way'' may not happen frequently, as Collingwood discusses in \emph{The Idea of History} \cite[pp.\ 41 f.]{Collingwood-IH}: \begin{quote} Greco-Roman humanism, however, had a special weakness of its own because of its inadequate moral or psychological insight. It was based on the idea of man as essentially a rational animal, by which I mean the doctrine that every individual human being is an animal capable of reason\lips Now the idea that every agent is wholly and directly responsible for everything that he does is a na\"{\i}ve idea which takes no account of certain important regions in moral experience\lips Looking back over our actions, or over any stretch of past history, we see that something has taken shape as the actions went on which certainly was not present to our minds, or to the mind of any one, when the actions which brought it into existence began. The ethical thought of the Greco-Roman world attributed far too much to the deliberate plan or policy of the agent, far too little to the force of a blind activity embarking on a course of action without foreseeing its end and being led to that end only through the necessary development of that course itself. \end{quote} Obviously \emph{we} do not know the full shape that our lives are going to take, no matter how deliberately we act. A simple example arises from when I started playing chess. At first, no matter how carefully I set out, intending not to make simple mistakes, I made them anyway. Only by experience could I learn not to make them. We may say that the shape that our lives will take must be present, if not to our own mind, then to \emph{some} mind, such as the mind of God. This would be a presupposition. It may be ridiculed as a fairy tale; but then another fairy tale may be substituted for it. If we end up doing things that we did not plan, we may say that, nonetheless, our Unconscious brought those things about, deliberately. This may be a useful presupposition, but we can make it only under strain, in the sense of page \pageref{strain}; for it conflicts with the observation (mentioned in note \ref{scepticism2}, page \pageref{scepticism2}) that we \emph{are} in control of what we do, or at least of what we have \emph{yet} to do or are \emph{about} to do. Oedipus may have been unable to falsify the Oracle; but he \emph{thought} that he could. A modern fictional example is in Herman Wouk's 1955 novel \emph{Marjorie Morningstar} \cite[ch.\ 42, pp.\ 483--489]{Wouk}, set in the 1930s. Here a faithful Freudian believes in an Unconscious controller of destiny---% at least the destiny of \emph{other} people. The strain of believing in it for \emph{himself} becomes intolerable when he kills his hated wife by accident (he falls asleep at the wheel).} %%%%%%%%%%%%%%%%%%%%%%%%%%%% That which causes may, as we have seen, come into operation through the activity of a second human agent.%%%%% \footnoteB{\label{operation}Collingwood does not say that that which causes may \emph{be} a second human agent, only that it may ``come into operation'' through a second agent. The meaning of this is given on page \pageref{come-into-op}. As originally described, starting on page \pageref{123}, a cause in sense II or III is an event in nature, or a state of things; no cause in sense I is described, but only that which is caused. Only in the next chapter, on page \pageref{quod}, do we see a \emph{cause} in sense I described: it is not simple, but made up of the ``cause that'' and ``cause for the sake of which'' (page \pageref{quod}; see also Appendix \ref{app:Aristo}). The ``cause that'' (\emph{causa quod}) is again a ``situation or state of things''; but it is emphatically not merely that, because the effect is not just an event (or situation, or state), but a deliberate \emph{act.} The cause \emph{that} the act happens must be known to the agent. The cause \emph{for the sake of which} the act happens (\emph{causa ut}) is the agent's intention. Both of these may ``come into operation''% ---may we say, ``be effected,'' or ``be caused''?---% by another agent.} %%%%%%%%%%%%%%%%%%%%%% In sense II that which is caused is an event in nature; but the word `cause' still expresses an idea relative to human conduct, because that which causes is something under human control, and this control serves as means whereby human beings can control that which is caused. In this sense, \uline{the cause of an event in nature is the handle}, so to speak, \uline{by which human beings can manipulate it}.%%%%% \footnoteB{Scholars have picked up on Collingwood's metaphorical use of ``handle,'' as I see in two commentaries \cite{Kronfeldner,Woodward} in the issue of the \emph{International Journal of Epidemiology} that reprints Collingwood's original article on cause \cite{Collingwood-IC}. Collingwood proposes no complementary kind of cause, whereby nature ``manipulates'' human beings. As recalled in note \ref{operation}, a cause in sense I, be it a ``cause that'' or ``cause for the sake of which,'' may be superficially a situation in nature; but it is this only \emph{as understood by the agent who is caused to do something.} Herodotus traces the origin of geometry (or surveying---\gr{it'ia} and its equivalent \gr{pr'ofasis} in, for example, the Hippocratic writings of the fifth century before Christ. \uline{A cause in sense II is} never able by itself to produce the corresponding effect. The switch, as I said, only works the light \uline{subject to certain} indispensable conditions. Among these are the existence of an appropriate current and its maintenance by insulation and contacts. These are called \uline{\emph{conditiones sine} \emph{quibus non}}. Their existence, over and above the cause, constitutes one of the differences between sense II and sense III of the word `cause'. As we shall see in the next chapter, a cause in sense III requires no such accompaniment. A cause in sense II is conditional, a cause in sense III is unconditional. This distinction was correctly understood \uline{by John Stuart Mill}, whose formal definition of the term `cause' is a definition of sense III, but who recognizes that ordinarily when people speak of a cause they are using the word in sense II. A cause, he tells us, is the invariable unconditional antecedent of its effect. This antecedent, he thinks, is always complex, and any one of the elements that go to make it up is called a condition. But \uline{what people ordinarily call a cause is one of these conditions, arbitrarily%%%%% \label{arbitrary} selected}, and dignified by a [\textbf{302}] mere abuse of language with a name that properly belongs to the whole set.%%%% \footnoteA{`Since then, mankind are accustomed with acknowledged propriety so far as the ordinances of language are concerned, to give the name of cause to almost any one of the conditions of a phenomenon, or any portion of the whole number, \emph{arbitrarily selected,} without excepting even those conditions which are purely negative, and in themselves incapable of causing anything; it will probably be admitted without longer discussion, that no one of the conditions has more claim to that title than another, and that \emph{the real cause of the phenomenon is the assemblage of all its conditions.}' (J. S. Mill, \emph{System of Logic,} Book III, chap.\ \textsc v, \S\ 3; ed.\ \textsc i, vol.\ i, p.\ 403, my italics.)} %%%%%%%%%%%%%%%%%%%%%%%% Mill deserves great credit for seeing that the word `cause' was used in these two different ways.%%%%% \footnoteB{\label{credit}% To assign credit is to imply \begin{inparaenum}[(1)] \item that one is qualified to judge who deserves it, and \item that the credit one has to offer is desirable. \end{inparaenum} See also note \ref{feudal}, page \pageref{feudal}, on feudal deference.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% But his account of the relation between a cause in sense II and the conditions that accompany it is not quite satisfactory. Closer inspection would have shown him that the `selection' of one condition to be dignified by the name of cause is by no means arbitrary. It is made according to a principle.%%%%% \footnoteB{Presumably an arbitrary selection is a \emph{capricious} selection: this is how the word ``arbitrary'' is used. Etymologically though, an arbitrary selection is a selection by an \emph{arbiter,} a judge. A true judge will use the principle that Collingwood refers to.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% The `condition' which I call the cause (in sense II) of an event in which I take a practical interest is the condition I am able to produce or prevent at will.%%%%% \footnoteB{The key word here is ``interest,'' which comes ultimately from the Latin verb \emph{interesse.} Skeat \cite{Skeat} derives our noun more precisely, via French, from the 3rd person singular form, \emph{interest,} of the Latin verb. As for the English verb ``interest,'' it was once ``interess,'' whose past participle ``interess'd'' became our form of the verb. All of this is corroborated, but not so clearly, and except for the last detail, by the more recent \emph{Oxford Concise Dictionary of English Etymology} \cite{CODEE}. Cause is something that we are \emph{interested} in finding. Originally interest is practical; then it becomes theoretical as well, by an ``historical process'' such as Collingwood discusses on page \pageref{cat} in connection with the word ``cat.''} %%%%%%%%%%%%%%%%%%%%%%% Thus, if my car fails to climb a steep hill, and I wonder why, I shall not consider my problem solved by a passer-by who tells me that the top of a hill is farther away from the earth's centre than its bottom, and that consequently more power is needed to take a car uphill than to take her along the level.%%%% \footnoteB{This kind of absurdity is a basis for jokes told today.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% All this is quite true; what the passerby has described is one of the conditions which together form the `real cause' (Mill's phrase; what I call the cause in sense III) of my car's stopping; and [\textbf{303}] as he has `arbitrarily selected' one of these and called it the cause, he has satisfied Mill's definition of what the word ordinarily means. But suppose an A.A. man comes along, opens the bonnet, holds up a loose high-tension lead, and says: `Look here, sir, you're running on three cylinders'. My problem is now solved. I know the cause of the stoppage. It is the cause, just because it has not been `arbitrarily selected'; it has been correctly identified as the thing that I can put right, after which the car will go properly. If I had been a person who could flatten out hills by stamping on them the passer-by would have been right to call my attention to the hill as the cause of the stoppage; not because the hill was a hill but because I was able to flatten it out. To be precise, \uline{the `condition' which is thus `selected' is in fact not `selected' at all}; for selection implies that the person selecting has before him a finite number of things from among which he takes his choice. But this does not happen. In the first place the conditions of any given event are quite possibly infinite in number, so that no one could thus marshal them for selection even if he tried. In the second place no one ever tries to enumerate them completely. Why should he? If I find that I can get a result by certain means I may be sure that I should not be getting it unless a great many conditions were fulfilled; but so long as I get it I do not mind what these conditions are. If owing to a change in one of them I fail to get it, I still do not want to know what they all are; I only want to know what the one is that has changed.%%%%% \footnoteB{Research is not divided into two stages, the first being to assemble all possible causes, and the second, to determine which of them is a condition that one can control. The whole course of research requires an awareness of what one can control. This theme is found in Collingwood's account of historical research in \emph{The Principles of History} \cite[pp.\ 36 f.]{Collingwood-PH}: \begin{quotation} \lips A sensible question (the only kind of question that a scientifically competent man will ask) is a question which you think you have or are going to have evidence for answering\lips It was a correct understanding of this truth that underlay Lord Acton's great precept, `Study problems, not periods.'\lips Scientific historians study problems: they ask questions, and if they are good historians they ask questions which they see their way to answering. It was a correct understanding of the same truth that led Monsieur Hercule Poirot to pour scorn on the `human blood-hound' who crawls about the floor trying to collect everything, no matter what, which might conceivably turn out to be a clue; and to insist that the secret of detection was to use what, with possibly wearisome iteration, he called 'the little grey cells'. You can't collect your evidence before you begin thinking, he meant: because thinking means asking questions (logicians, please note), and nothing is evidence except in relation to some definite question. \end{quotation} Collingwood was evidently a fan of detective fiction; in \emph{The Principles of Art} he spoke of the fun it would be to take a break from writing that book, to lie in the garden reading Dorothy Sayers \cite[p.\ 95]{Collingwood-PA}. This mention of Sayers is not the one in the book's Index.} [\textbf{304}] From this a principle follows which I shall call `the relativity of causes'.% \index{relativity of causes}\label{rel-cause}% \footnoteB{This principle is mentioned several times, through the beginning of the next chapter on page \pageref{C-int-ref}.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Suppose that the conditions of an event \emph y include three things, \gr a, \gr b, \gr g; and suppose that there are three persons A, B, C, of whom A is able to produce or prevent \gr a and only \gr a; B is able to produce or prevent \gr b and only \gr b; and C is able to produce or prevent \gr g and only \gr g. Then if each of them asks `What was the cause of \emph y?' each will have to give a different answer. For A, \gr a is the cause; for B, \gr b; and for C, \gr g. The principle may be stated by saying that \emph{for any given person the cause in sense II of a given thing is that one of its conditions which he is able to produce or prevent.}%%%%% \footnoteB{Surely the person might be able to produce or prevent more than one of the conditions.} %%%%%%%%%%%%%%%%%%% For example, a car skids while cornering at a certain point, strikes the kerb, and turns turtle. From the car-driver's point of view the cause of the accident was cornering too fast, and the lesson is that one must drive more carefully. From the county surveyor's point of view the cause was a defect in the surface or camber of the road. and the lesson is that greater care must be taken to make roads skid-proof. From the motor-manufacturer's point of view the cause was defective design in the car, and the lesson is that one must place the centre of gravity lower. If the three parties concerned take these three lessons respectively to heart accidents will become rarer.%%%%% \footnoteB{Or if \emph{any one of them} takes his own lesson to heart, then accidents will become rarer. But if each of the parties is interested merely in taking his or her own ease, then nothing will happen, because each will put the responsibility on the other persons.} %%%%%%%%%%%%%%%%%%% A knowledge of the causes of accidents will be gained in such a sense that knowledge is power: causes are causes in sense II, and knowledge of the cause of a thing we wish to prevent is (not merely brings, but is) knowledge how to prevent it. As in [\textbf{305}] the science of medicine%%%%% \footnoteC{I would put a comma here, but the original text has none.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% so in the study of `accidents', where `accident' means something people wish to prevent, the word `cause' is used in sense II. As in medicine, therefore, so in the study of `accidents' the use of the word in any other sense, or its use by some one who fails to grasp the implications of this sense, leads to confusion.%%%%% \footnoteB{Here is the \emph{practical} reason to distinguish senses of the word ``cause.''} %%%%%%%%%%%%%%%%%%%%%%%%%% If the driver, the surveyor, and the manufacturer agreed in thinking they knew the cause of the accident I have described, but differed as to what it was, and if each thought that it was a thing one of the others could produce or prevent, but not himself, the result would be that none of them would do anything towards preventing such accidents in future, and their so-called knowledge of the cause of such accidents would be a `knowledge' that was not, and did not even bring, power.%%%%% \footnoteB{Except the power of complaining about the negligence of other people.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%% But since in the present context the word `cause' is used in sense II, the reason why their `knowledge' of the `cause' of such accidents does not enable them to prevent such accidents is that it is not knowledge of their cause. What each of them mistakes for such knowledge is the following nonsense proposition: `the cause of accidents like this is something which somebody else is able to produce or prevent, but I am not.' Nonsense, because `cause' means `cause in sense II', and owing to the relativity of causes% \index{relativity of causes} `the cause of this accident' means `that one of its conditions which I am able to produce or prevent'.%%%%%% \footnoteB{Collingwood has allowed that there may be other senses of the word ``cause'' than the three that he has enumerated. Why should the word as used in the ``nonsense proposition'' be interpreted in such a way as to make the proposition nonsense? Perhaps because Collingwood is a practical man. He is responding to persons who would seem to be trying to avoid taking action.} %%%%%%%%%%%%%%%%%%%%%%%%%%% Hence the folly of blaming other people in respect of an event in which we and they are together involved. Every one knows that such blame is foolish; [\textbf{306}] but without such an analysis of the idea of causation as I am here giving it is not easy to say why.%%%%% \footnoteB{More needs to be said here. I do not suppose Collingwood is recommending stoicism. It is foolish to blame others for one's own problems; and yet sometimes complaints have an effect. Istanbul's Gezi Park would be gone now, if people had not complained about its planned destruction.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% In medicine the principle of the relativity of causes% \index{relativity of causes} means that, since any significant statement about the cause of a disease is a statement about the way in which that disease can be treated, two persons who can treat the same disease in two different ways will make different statements as to its cause. Suppose that one medical man can cure a certain disease by administering drugs, and another by `psychological' treatment. For the first the `cause' of the disease will be definable in terms of bio-chemistry; for the second in terms of psychology. If the disease itself is defined in terms of bio-chemistry, or in terms that admit of explanation or analysis in bio-chemical language, the definition of its cause in terms of psychology may be thought to imply an `interactionist' theory of the relation between body and mind; and may be thought objectionable in so far as such theories are open to objection. But this would be a mistake.%%%%% \footnoteB{\label{NL-interact}Collingwood himself ridicules and refutes ``Psycho-physical Interactionism'' in Chapter II, ``The Relation Between Body and Mind,'' of \emph{The New Leviathan} \cite[p.\ 10]{Collingwood-NL}, by pointing out that no physicist will accept it. The philosopher is not allowed to tell the practical person (the physicist who is getting results) what she can do. Collingwood argues likewise now; the philopher who (however rightly) rejects ``interactionism'' cannot tell the the medical doctor what words to use to describe effective treatments. It is for the philosopher to understand what the words \emph{mean.} See page \pageref{word-choice} and note \ref{word-choice}.} %%%%%%%%%%%%%%%%%%%%%% Definition of its cause in terms of psychology implies no theory as to the relation between body and mind. It simply records the fact that cases of the disease have been successfully treated by psychological methods, together with the hope that psychological methods may prove beneficial in future cases. To speak of this as `evidence for an interactionist theory' would be to talk nonsense. A corollary of the relativity principle% \index{relativity of causes} is that \emph{for a person who is not able to produce or prevent any of its conditions a given event has no cause in sense II at all}, [\textbf{307}] and any statement he makes as to its cause in this sense of the word will be a nonsense statement. Thus the managing director of a large insurance company once told me that his wide experience of motor accidents had convinced him that the cause of all accidents was people driving too fast. This was a nonsense statement; but one could expect nothing better from a man whose practical concern with these affairs was limited to paying for them.%%%%% \footnoteB{Are such supercilious statements a reason why Collingwood is little read? Do they put people off, as Blackburn suggested in the quotation on page \pageref{off-putting}? Collingwood has a good point: One cannot---and therefore should not try to---% pass judgment on matters that one cannot influence. But then is the insurance company director under Collingwood's influence? In any case, the director's job is not exactly to pay for accidents, but to avoid paying for them. He may be able to do this by advising people on how to drive better. Possibly Collingwood felt that the director's ``advice'' was not being offered in the correct spirit. Or perhaps he understood the advice was an unintended joke, since every car accident can be blamed on excessive speed. Two stationary vehicles will never collide.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% In sense II of the word `cause' only a person who is concerned with producing or preventing a certain kind of event can form an opinion about its cause. For a mere spectator there are no causes.%%%%% \footnoteB{And yet spectators are always telling other people what they ought to do, or ought to have done.} %%%%%%%%%%%%%%%%%%%%%%%%%%% When Hume tried to explain how the mere act of spectation could in time generate the idea of a cause, where `cause' meant the cause of empirical science, that is, the cause in sense II, he was trying to explain how something happens which in fact does not happen.%%%%% \footnoteB{Collingwood refers to what happens, or not, ``in fact''; but he is speaking of a \emph{mere} spectator, and strictly speaking there is no such person. Everybody has experience of \emph{doing} as well as \emph{watching,} and this is why they can complain about what other people fail to do: from their own experience, they think they themselves could do it. See note \ref{tries}.} %%%%%%%%%%%%%%%%%%%%%%%%%% If sciences are constructed consisting of causal propositions in sense II of the word `cause', they will of course be in essence codifications of the various ways in which the people who construct them can bend nature to their purposes, and of the means by which in each case this can be done. Their constituent propositions will be (\emph a) experimental, (\emph b) general. (\emph a) In calling them experimental I mean that they will be established by means of experiment. No amount of observation will serve to establish such a proposition; for any such proposition is a declaration of ability to produce or prevent a certain state of things by the use of certain means; and no one knows [\textbf{308}] what he can do, or how he can do it, until he tries.%%%%% \footnoteB{\label{tries}Nobody knows what she can do until she tries. A key point.} %%%%%%%%%%%%%%%%%%%%%%%%%% By observing and thinking he may form the opinion that he can probably do a given thing that resembles one he has done in the past; he may, that is, form an opinion as to its cause;%%%%% \footnoteB{\label{thought-experiment}% This way of forming an opinion might be what is called ``thought experiment'' (see also note \ref{te-2}, page \pageref{te-2}). In any case, it is not based on \emph{mere} thought, but thought as based on past experience, as Collingwood seems to acknowledge. But then \emph{all} thought would seem to be based on \emph{past} experience. And yet we can distinguish past experience from experience that is being kept current through continual exercise and renewal.} %%%%%%%%%%%%%%%%%%%%%%%%%%% but he cannot acquire knowledge. (\emph b) Because the proposition `\emph x causes \emph y', in sense II of the word `cause', is a constituent part of a practical science, it is essentially something that can be applied to cases arising in practice; that is to say, the terms \emph x and \emph y are not individuals but universals, and the proposition itself, rightly understood, reads `any instance of \emph x is a thing whose production or prevention is means respectively of producing or preventing some instance of \emph y'. It would be nonsense, in this sense of the word `cause', to inquire after the cause of any individual thing as such.%%%%% \footnoteB{\label{RP-Granted}% \emph{Blame} or \emph{responsibility} is assigned in individual cases. But here we are alluding to sense I of ``cause.'' A state of affairs cannot be to blame for anything, but the humans who created the state of affairs are to blame. Metaphorically though, we may blame the damn car for the accident. Is this another remnant of animism (note \ref{animist}, page \pageref{animist})? Collingwood has pointed out (page \pageref{metaphor}) that metaphorical uses of words are still proper. Now he seems to accepts the distinction between metaphorical and literal uses. In \emph{Religion and Philosophy}; Part III, ``From Metaphysics to Theology''; Chapter III, ``Miracle''; \S3, (\emph b), ii.\ (the last division of the book), he would seem to dispense with the distinction \cite[p.\ 212]{Collingwood-RP}: \begin{quote} Granted---and by now we seem bound to grant---that a ball, let drop, falls in virtue not of an inexorable law but of a volition, and that the volition might will otherwise, we may still say that the possibility of a ball's thus changing its habits\label{habit-2} need not seriously disturb our practical calculations. We have to deal not only with things, but with men; and if the engineer feels justified in calculating the strength of his materials on a basis of absolute uniformity, the organiser of labour is no less ready to calculate the average output of a workman and to act on his calculations. If we try to carry the principle of uniformity too far, it will fail us whether our assumption is that any man will write an equally good epic or that any steel will make an equally good razor. \end{quote} Then, everything was alive; now, Collingwood is more practical---% in particular, more concerned with the current state of science as it is understood by its practitioners.} %%%%%%%%%%%%%%%%%%%%%%%%% \uline{It is a peculiarity of sense II that every causal proposition is a general proposition or `propositional function'}. In sense I every causal proposition is an individual proposition. \label{indiv-gen}%%%%% In sense III causal propositions might equally well be either individual or general.%%%%% \footnoteB{I quote this sentence in note \ref{indiv-gen-cite}, page \pageref{indiv-gen-cite}.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% If the above analysis%%%%% \label{means-end} %%%%%%%%%%%%%%%%% of the cause-effect relation (in sense II) into a means-end relation is correct, why do people describe this means-end relation in cause-effect terminology? People do not choose words at random; they choose them because they think them appropriate.%%%%% \footnoteB{\label{word-choice}Right; but sometimes, it would seem, they think incorrectly.} %%%%%%% If they apply cause-effect terminology to things whose relation is really that of means and end the reason must be that they want to apply to those things some idea which is conveyed by the [\textbf{309}] cause-effect terminology and not by the means-end terminology. What is this idea? The answer is not doubtful. \uline{The cause-effect terminology conveys an idea} not only of one thing's leading to another but \uline{of one thing's forcing another to happen or exist; an idea of power or compulsion%%% \label{compulsion} or constraint}. From what impression, as Hume asks, is this idea derived? I answer, from impressions received in our social life, in the practical relations of man to man; specifically, from the impression of causing (in sense I) some other man to do something when, by argument or command or threat or the like, we place him in a situation in which he can only carry out his intentions by doing that thing; and conversely, from the impression of being caused to do something. Why, then, did people think it appropriate to apply this idea to the case of actions in which we achieve our ends by means, not of other human beings, but of things in nature? \uline{Sense II of the word `cause' is especially a Greek sense}; in modern times it is especially associated with the survival or revival of Greek ideas in the earlier Renaissance thinkers; and \uline{both the Greeks and the earlier Renaissance thinkers held quite seriously an animistic theory of nature}. They thought of what we call the material or physical world as a living organism or complex of living organisms, each with its own sensations and desires and intentions and thoughts. In Plato's \emph{Timaeus}, and in the Renaissance Platonists whose part in the formation of modern natural science was so decisive, the constant use of language with [\textbf{310}] animistic implications is neither an accident nor a metaphor; these expressions are meant to be taken literally and to imply what they seem to imply, namely that the way in which men use what we nowadays call inorganic nature as means to our ends is not in principle different from the way in which we use other men. We use other men by assuming them to be free agents with wills of their own, and influencing them in such a way that they shall decide to do what is in conformity with our plans. This is `causing' them so to act in sense I of the word `cause'. If `inorganic nature' is alive in much the same way as human beings, we must use it according to much the same principles; and therefore we can apply to this use of it the same word `cause', as implying that there are certain ways in which natural things behave if left to themselves, but that man, being more powerful than they, is able to thwart their inclination to behave in these ways and make them behave not as they like but as he likes. To sum up. \uline{Sense II of the word `cause' rests on} two different ideas about the relation between man and nature.%%%%% \footnoteB{\label{ideas}% In Chapter XXXIII, ``Causation in Kantian Philosophy,'' page \pageref{dilemmas}, we shall see the anthropocentric and anthropomorphic \emph{dilemmas,} constituting the insecure foundation of nineteenth-century science: do we still use cause in sense II?} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 1. \uline{The anthropocentric idea} that man looks at nature from his own point of view; not the point of view of a thinker, anxious to find out the truth about nature as it is in itself, but the point of view of a practical agent, anxious to find out how he can manipulate nature for the achieving of his own ends. 2. \uline{The anthropomorphic idea} that man's manipulation of nature resembles one man's manipulation [\textbf{311}] of another man, because natural things are alive in much the same way in which men are alive, and have therefore to be similarly handled. The first idea is admittedly part of what civilized and educated European men nowadays think about their relations with nature. The second idea is part of what they notoriously did think down to (say) four centuries ago. How they began to get rid of this idea, and how completely they have even now got rid of it, are questions I shall not raise. My point is that even to-day, when they use the word `cause' in sense Il, they are talking as if they had not yet entirely got rid of it. For if the vocabulary of practical natural science were overhauled with a view to eliminating all traces of anthropomorphism, language about causes in sense II would disappear and language about means and ends would take its place. Fifty years ago, anthropologists were content to note the fact that `survivals'%%%%% \label{survivals} occur. Since then, they have seen that the occurrence of such things constitutes a problem, and a difficult one.%%%%% \footnoteB{Collingwood's interest in anthropology is shown in \emph{The Principles of Art} \cite{Collingwood-PA} and the postumous \emph{Philosophy of Enchantment} \cite{Collingwood-PE}.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% `Students have made some progress in ascertaining what causes folklore to decay, but what causes the surviving elements to survive? What vacuum does the survival fill?\lips These questions\lips remain a problem for the future.'%%%%% \footnoteA{Charlotte S. Burne, \emph{Folklore,} vol.\ xxii (1911), p.\ 37.} %%%%%%%%%%%%%%%%%%%%% What causes the survival of language which taken literally implies the survival of supposedly obsolete thought-forms is, I submit, the fact that \uline{these thought-forms are not so dead as they are supposed to be}. It is certainly true that modern [\textbf{312}] natural science has tried very hard to expel anthropomorphic elements from its conception of nature. Among natural scientists to-day it is orthodox to take the will for the deed. For the historical metaphysician it is a question how far this anti-anthropomorphic movement has been successful. The continued use of the word `cause' in sense II is prima-facie evidence that its success has not been complete. %[\textbf{313}] \mychap[XXXII. Causation in Theoretical Natural Science]{XXXII\\ CAUSATION IN THEORETICAL NATURAL SCIENCE}%XXXII \textsc{Sense} III of the word `cause' represents an attempt to apply it not to a `practical' but to a `theoretical' science of nature. I shall first explain the characteristics which would belong to this sense if the attempt were successful, and then consider certain difficulties which in the long run prove fatal to it.%%%%%%%%%%%%% \footnoteB{\label{RP-Popular}% Collingwood does this---explain characteristics and consider difficulties---also in the subsection of \emph{Religion and Philosophy} on ``The paradox of causation'' already cited (notes \ref{RP} and \ref{RP-whole}, pages \pageref{RP} and \pageref{RP-whole}). He says at the head of the chapter, ``Matter'' \cite[p.\ 72]{Collingwood-RP}, \begin{quote} \textsc{Popular} metaphysic distinguishes two categories of reality, mind and matter\lips Matter is thus subject to the law of causation, the law that whatever happens has a cause, external to itself, which determines it to happen in this way and in no other. This law of causation does not apply to mind, whose changes of state are initiated freely from within, in the form of acts of will. These acts of will may influence matter, but they cannot alter or in any way affect the operation of the laws which govern the movements of matter. \end{quote} The cause considered is only of sense III. Kant's ``law of causation'' is considered in the next chapter (page \pageref{ch33}).} %%%%%%%%%%%%%%%%%%%%%%% In the contingent world to which sense II belongs a cause is contingent (\emph a) in its existence, as depending for its existence on human volition, (\emph b) in its operation, as depending for the production of its effect on \emph{conditiones sine quibus non}. \uline{In the necessary world to which sense III belongs a cause is necessary} (\emph a) \uline{in its existence}, as existing whether or no human beings want it to exist, (\emph b) \uline{in its operation}, as producing its effect no matter what else exists or does not exist. There are no \emph{conditiones sine quibus non}. The cause leads to its effect by itself, or `unconditionally'; in other words the relation between cause and effect is a one-one relation. There can be no relativity of causes,% \index{relativity of causes} and no diversity of effects due to fulfilment or nonfulfilment of conditions.%% \footnoteB{\label{RP-tree}% Collingwood reasons this out in \emph{Religion and Philosophy} \cite[pp.\ 83 f.]{Collingwood-RP}, in (i) of the subsection already cited (see note \ref{RP}, page \pageref{RP}): \begin{quotation} If we search for the particular cause of a given particular effect, we shall find this cause to be invariably complex, even when it is often described as simple. Thus, the gale last night blew down a tree in the garden. But it would not have done so except for many other circumstances. We must take into account the strength of the tree's roots, its own weight, the direction of the wind, and so on. If some one asks, ``why did the tree fall?'' we cannot give as the right and sufficient answer, ``because of the wind.'' We might equally well give a whole series of other answers: ``because the wind was in the north-west''; ``because the tree had its leaves on''; ``because I had not propped it''; and so on. Each of these answers is a real answer to the question, but none of them is the only answer or the most right answer. No one of them can claim to give the cause in a sense in which the others do not give the cause. Is there then, we may ask, such a thing as \emph{the} cause at all? is there not simply a number of causes? No, there does seem to be one cause and no more; but that cause is not one simple event but a large, indeed an infinitely large, number of events and conditions all converging to the one result. \end{quotation} The conclusion is, ``The only real cause seems to be a total state of the universe.'' Collingwood will presently go through a similar chain of reasoning concerning action at a distance.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% I propose to distinguish the one-many and many-one%%%%% \footnoteA{One-many, because a cause in sense II leads to its effect only when the \emph{conditiones sine quibus non} are fulfilled. Many-one, because of the relativity of causes% \index{relativity of causes}\label{C-int-ref} (see p.\ 304 [our page \pageref{rel-cause}]).%%% \footnoteB{I do not know why the meanings of ``one-many'' and ``many-one'' should be assigned thus, and not the other way around. Only one of the terms will be used again, and only once, on page \pageref{one-many}.} %%%%%%%%%%% } %%%%%%%%%%%%%%%%%%%%%%% character of the cause-effect relation in sense II [\textbf{314}] from its one-one character in sense III by calling these senses \emph{loose} and \emph{tight} respectively. A loose cause requires some third thing other than itself and its effect to bind the two together, namely a group of \emph{conditiones sine quibus non}; a tight cause is one whose connexion with its effect is independent of such adventitious aids. In\label{314} order to illustrate the implications of sense III, I will refer to the contradiction between the traditional denial of \emph{actio in distans} (which, I suppose, would hold as against action across a lapse of time no less than across a distance in space) and the assumption, commonly made nowadays, that a cause precedes its effect in time. I shall argue that \uline{\emph{actio in distans} is perfectly intelligible in sense II but nonsense in sense III}. If I set fire to one end of a time-fuse, and five minutes later the charge at its other end explodes, there is said to be a causal connexion between the first and second events, and a time~interval of five minutes between them. But this interval is occupied by the burning of the fuse at a determinate rate of feet per minute; and this process is a \emph{conditio sine qua non} of the causal efficacy ascribed to the first event. That is to say, the connexion between the lighting of the fuse and the detonation of the charge is causal in the loose sense, not the tight one. If in the proposition `\emph x causes the explosion' we wish to use the word `cause' in the tight sense, \emph x must be so defined as to include in itself every such \emph{conditio sine qua non}. It must include the burning of the whole fuse; not its burning until `just [\textbf{315}] before' that process reaches the detonator, for then there would still be an interval to be bridged, but its burning until the detonator is reached. Only then is the cause in sense III complete; and when it is complete it produces its effect, not afterwards (however soon afterwards) but then. Cause in sense III is simultaneous with effect. Similarly, it is coincident with its effect in space. The cause of the explosion is where the explosion is. For suppose \emph x causes \emph y, and suppose that \emph x is in a position \emph p\textsubscript 1 and \emph y in a position \emph p\textsubscript 2, the distance from \emph p\textsubscript 1 to \emph p\textsubscript 2 being \gr d. If `cause' is used in sense II, \gr d may be any distance, so long as it is bridged by a series of events which are \emph{conditiones sine quibus non} of \emph x causing \emph y. But if `cause' is used in sense III, \gr d must = 0. For if it did not, \emph p\textsubscript 2 would be any position on the surface of a sphere whose centre was \emph p\textsubscript 1 and whose radius would = \gr d; so the relation between \emph p\textsubscript 1 and \emph p\textsubscript 2 would be a one-many%%%% \label{one-many} %%%%%%%%%%%%% relation. But the relation between \emph x and \emph y, where \emph x causes \emph y in sense III, is a one-one relation. Therefore, where \gr d does not = 0, \emph x cannot cause \emph y in sense III.%%%%% \footnoteB{The technical symbols obscure the point, which seems to be this: If lighting a fuse causes an explosion ten feet away, then it ought to cause an explosion at every point that is ten feet away, unless there are some additional \emph{conditiones sine quibus non}; and this shows that ``cause'' is being used in sense II.} %%%%%%%%%%%%%%%%%% The denial of \emph{actio in distans}, spatial or temporal, where the `agent' is a cause in sense III, is therefore not a `prejudice'%%%%% \footnoteA{As Russell calls it: \emph{Mysticism and Logic,} cit., p.\ 192.} %%%%%%%%%%%%%%%%%%%%%%%% but is logically involved in the definition of sense III.%%%% \footnoteB{Collingwood's logic must not be universally recognized, if its result can be considered a prejudice.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \uline{The main difficulty about sense III is to explain what is meant by saying that a cause `produces' or `necessitates' its effect}. When similar language is used of senses I and II we know what it means. In sense I [\textbf{316}] it means that \emph x affords somebody a motive for doing \emph y; in sense II, that \emph x is somebody's means of bringing \emph y about.%%%%% \footnoteB{In short, heretofore, instances of producing or necessitating involved a person, an agent.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% But what (since it cannot mean either of these) does it mean in sense III? There are two well-known answers to this question, which may be called the rationalist and empiricist answers respectively. (i) \uline{The rationalist answer runs: `necessitation means impli\-% cation'}.\label{implication}%%%%% \footnoteB{\label{cause-not-imp}% This is wrong, because mathematical deduction is simply not how we draw conclusions about the physical world. Read on and see note \ref{cause-not-imp-2}.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% A cause, on this view, is a `ground', and its relation to its effect is the relation of ground to consequent, a logical relation. When some one says that \emph x necessitates \emph y he means on this view that \emph x implies \emph y, and is claiming the same kind of insight into \emph y which one has (for example) into the length of one side of a triangle given the lengths of the other two sides and the included angle.%%%%%%%%%%%%%%%5 \footnoteB{``Synthetically,'' the length is determined by what Euclid proves as Proposition 4 of Book \textsc i of the \emph{Elements} \cite{MR17:814b,Euclid-homepage,MR1932864}, according to the principle suggested in one of the Common Notions, that equal segments and angles are congruent. ``Analytically,'' the length is determined by the Law of Cosines, \begin{math} a^2=b^2+c^2-2bc\cos\alpha \end{math}, where $a$, $b$, and $c$ are the sides of the triangle, and $\alpha$ is the angle subtended by $a$.} %%%%%%%%%%%%%%%%%%%%%%%%%% Whatever view one takes as to the nature of implication, one must admit that in such a case the length of the third side can be ascertained without measuring it and even without seeing it, e.g.\ when it lies on the other side of a hill.%%%%% \footnoteB{\label{180}The proposed logical deduction is based on the assumption (or presupposition)%%%%% \footnoteD{\label{assume}In Chapter IV, ``On Presupposing,'' of the \emph{Essay,} on the original page 27, Collingwood makes the definition, ``\emph{To assume is to suppose by an act of free choice.}'' The reason to make the distinction is that some suppositions are not chosen, either because one is unaware of them, or because one is unaware of alternatives.} %%%%%%%%%%%%%%%%%%%%%% that the world is indeed Euclidean, so that the Law of Cosines holds. We are (and were in Collingwood's day) aware of other possibilities. The determination of the \emph{correct} possibility would seem to be based on observation. In each triangle heretofore measured, the angles are equal to two right angles, as is predicted by Proposition 32 of Book I of Euclid's \emph{Elements} \cite{MR17:814b}. We assume then that the same will be true for every triangle. However, we have no \emph{reason} to make this assumption, beyond a conviction that the world ought to have uniform unchanging curvature---% and therefore, according to our observations, zero curvature. Even if we question this conviction, there is a deeper one that remains: the world \emph{is} geometrical, in the current sense: it is a \emph{manifold,} it is ``locally Euclidean.''} %%%%%%%%%%%%%%%%%%% The implication theory, therefore, implies that `if the cause is given the effect follows', not only in the sense that whenever the cause actually exists the effect actually follows, but that from the thought of the cause the thought of the effect follows logically. That is to say, any one who wishes to discover the effect of a given thing \emph x can discover the answer by simply thinking out the logical implications of \emph x. Nothing in the nature of observation or experiment is needed.%%%%% \footnoteB{Except the observations mentioned in the previous note (note \ref{180}). And those observations do not prove conclusively that the world is Euclidean; so one should continue to measure the angles of triangles, just to make sure. We can understand the implication theory to be that there are axioms of the world that logically determine its progress; but the axioms must be found by observation. However, this may not be enough for a true ``implication theory,'' which might require us to understand the world entirely \emph{a priori.} See Collingwood's next paragraph.} %%%%%%%%%%%%%%%%%%%%%%%%%%%% This is in itself a tenable position in the sense that, if any one wants to construct a system of science in [\textbf{317}] which the search for causes means a search for grounds, there is nothing to prevent him from trying.%%%%% \footnoteB{\label{te-2}% And how is this search to be carried out: by experiment, or by the kind of ``thought experiment'' mentioned in note \ref{thought-experiment}, page \pageref{thought-experiment}?} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% This was in fact what Descartes tried to do. His projected `universal science' was to be a system of grounds and consequents. And if, as is sometimes said, modern physics represents a return in some degree to the Cartesian project, it would seem that the attempt is being made once more. But \uline{the rationalist theory of causation}, however valuable it may be as the manifesto of a particular scientific enterprise, \uline{cannot be regarded as an `analysis' of the causal propositions asserted by natural science} as it has existed for the last few centuries. If it were accepted, these propositions would have to be abandoned as untrue. \uline{For no one believes that they can be established by sheer `thinking'}, that is, by finding the so-called effects to be logically implied in the so-called causes.%%%% \footnoteB{\label{string}% It appears that string-theorists do try to work out the underlying nature of the world by sheer thinking: \begin{quote} Disagreement and controversy are of course necessary for science to progress, but there is always supposed to be a way to resolve a dispute by means of experiment or mathematics. In the case of string theory, however, this mechanism seems to have broken down. Many adherents and critics of string theory are so confirmed in their views that it is difficult to have a cordial discussion on the issue, even among friends. ``How can you not see the beauty of the theory? How could a theory do all this and not be true?'' say the string theorists. This provokes a heated response from skeptics: ``Have you lost your mind? How can you believe so strongly in \emph{any} theory in the complete absence of experimental test? \lips'' \end{quote} This Lee Smolin in \emph{The Trouble With Physics} \cite[p.\ xviii]{Smolin}. See also page \pageref{string-2} and note \ref{string-2} on conflict in physics.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% It is just because this is impossible that the questions what causes a given effect and what effect a given cause produces have to be answered by observation and experiment. Hence the result of establishing a science of the Cartesian type would be not an analysis of propositions of the type `\emph x causes \emph y' into propositions of the type `\emph x implies \emph y' but the disuse of causal propositions in that kind of science and the use of implicational propositions instead;%%%% \footnoteB{\label{cause-not-imp-2}% These ``implicational propositions'' may continue to use the language of causation. Our present aim is to understand what people \emph{mean} by the words that they use. Causation simply does not normally mean logical implication, regardless of whether one decides that instances of what is normally called causation are ``really'' instances of logical implication. If a schoolchild says, ``I don't want no Brussels sprouts!'' and a teacher tries to correct the grammar by saying, ``The double negative means you \emph{do} want Brussels sprouts,'' the teacher is wrong, regardless of the theory of grammar taught be the school.} %%%%%%%%%%%%%%%%%%%%%%%%% while in the sciences of observation and experiment causal propositions not analysable into implicational propositions would still be used; the meaning of `necessity' in these causal propositions being still doubtful.%%%%% \footnoteB{Collingwood's conclusion (on page \pageref{mod-phys}) is going to be that modern physics ``has eliminated the notion of cause altogether.'' And yet physics continues be a ``science of observation and experiment.'' Still, there are university departments of physics where (as far as I know) no experiments are actually done; it might be said that these departments are doing ``science of the Cartesian type,'' though presumably their work is based on experiments that have been done elsewhere. And I suppose that actual experimenters will be \emph{thinking} in terms of cause and effect.} %%%%%%%%%%%%%%%%%%%%%%%%%%% [\textbf{318}] This situation would not be illuminated by alleging that the sciences in which causal propositions occur are `backward' or `immature sciences'. Such a statement would imply that the idea of causation is a half-baked idea which when properly thought out will turn into the different idea of implication. This I take to be the Hegelian theory of the dialectic of concepts, and if any one wishes to maintain it I do not want to forbid him; but I must observe that it does not excuse him from answering the question what the half-baked idea is \emph{an sich}, that is, before its expected transformation has happened.%%%%% \footnoteB{\label{avoid}Collingwood gives his own answer, starting on page \pageref{most}: causation means compulsion in a human or ``anthropomorphic'' sense, and this meaning is a remnant of animist, Neoplatonic theology. Meanwhile, Collingwood has been showing, and will continue to show, that other attempts to explain causation are really attempts to \emph{avoid} explaining it. Collingwood's purpose can be seen, even in the names he has assigned to two attempted explanations: rationalist and empiricist. He will assign the name ``functionalist'' to a third attempt. These are all names of attempts to explain the \emph{world}: and since there is causation in the world, an explanation of the world will explain causation. But it is a \emph{presupposition} that there is causation in the world, a presupposition that is apparently jettisoned by rationalism and empiricism, even though these schools may possibly retain the \emph{terminology} of causation. But what does causation mean, \emph{before} one becomes a rationalist or empiricist? This is the question. It is an historical question. It is \emph{a priori} possible that one was a rationalist or empiricist all along; but Collingwood is showing that one in fact was not.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% (ii) I turn to \uline{the empiricist answer: `necessitation means ob\-% served uniformity of conjunction'. Like the former answer this one cannot be taken literally}; for no one, I think, will pretend that the proposition `\emph x necessitates \emph y' means merely `all the observed \emph x's have been observed to be conjoined with \emph y's', and does not also mean `\emph x's' observed in the future will also be conjoined with \emph y's'. In fact the question (so urgent for, e.g., Hume and Mill) how we proceed from the mere experience of conjunction to the assertion of causal connexion resolves itself into the question how we pass from the first of these to the second.%%%% \footnoteB{If Collingwood's assertion is not tautological, the question of how conjunction becomes causation ``resolves itself'' into the question of how, in the statement ``the observed $x$ have been conjoined with the $y$,'' the auxiliary verb ``have been'' becomes ``will be.''} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% For Hume and Mill the proposition `all the observed \emph x's have been observed to be conjoined with \emph y's' is not what we mean by saying `\emph x necessitates \emph y', it is only the empirical evidence on the strength of which we assert the very different proposition `\emph x necessitates \emph y'. Thus, if any one says `necessitation means observed uniformity of conjunction', it must be sup- [\textbf{319}] posed either that he is talking without thinking; or that he is carelessly expressing what, expressed more accurately, would run: `necessitation is something we assert on the strength of observed uniformity of conjunction', without telling us what he thinks necessitation to be; or, thirdly, that he is expressing still more carelessly what should run: `in order to assert a necessitation we must pass from the first of the above propositions to the second; now I cannot see how this is possible; therefore I submit that we ought never to assert necessitations, but on the occasions when we do assert them we ought to be asserting something quite different, namely observed conjunction'. Necessitation being again left undefined.%%%%% \footnoteB{Undefined, not in the way a geometer might leave terms like ``point'' and ``line'' undefined, but because, as in the last note (note \ref{avoid}), the empiricist is \emph{avoiding} an explanation of necessitation or causation.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% (iii) A third answer to our question has been given by Earl \uline{Russell}, in a paper%%%%% \footnoteA{`On the Notion of Cause', referred to on p.\ 69, above [not in the present extract].} %%%%%%%%%%%%%%%%%%%% of very great importance, to which I have already referred; but I want here and now to express my great admiration for it and my great indebtedness to it.%%%%% \footnoteB{\label{feudal}% To say that one is indebted is to imply that one can repay the debt, at least in theory. It puts the debtor on a level with the lender, or at least it connects the two. And in the present case the lender has been given his aristocratic title. I detect a \emph{survival} (in the sense of page \pageref{survivals}) of feudalism. Note also the ``reluctance,'' later in the paragraph, to traverse the lord. Perhaps this is all standard etiquette. See note \ref{credit}, page \pageref{credit}, on credit to Mill, but also Collingwood's approving quotation of Russell's republican sentiments on page \pageref{bitter}: ``The law of causality\lips is a relic of a bygone age, surviving, like the monarchy, only because it is erroneously supposed to do no harm.''} %%%%%%%%%%%%%%%%%%%%% He \uline{says: `\emph{necessary} is a predicate of a propositional function meaning that it is true for all possible values of its argument or arguments'}. This I will call the `functional' answer. In so far as it amounts to saying that causation in sense III implies a one-one relation between cause and effect, I entirely agree. But I find myself, very reluctantly, unable to accept all of what I take Earl Russell to mean. I will give two examples. (\gr a) \uline{How}, on the functional theory, \uline{could any one ever know} a causal proposition to be true,%%%%% \footnoteB{By the account in Chapter IV, ``On Presupposing,'' of the \emph{Essay} (mentioned also in note \ref{assume} on note \ref{180}, page \pageref{180}), a \emph{proposition} is true or false, because it is the answer to a question; an \emph{absolute presupposition,} not being the answer to a question, is not true or false, and so it is not a proposition.} %%%%%%%%%%%%%%%%%% or even know that the facts in his possession tended to justify [\textbf{320}] a belief in it? Only, so far as I can see, if there is a relation of implication between \emph x and \emph y. For `all \emph{possible} values' of \emph x may be an infinite number; and, even if they are not, it may not be practicable to examine them individually. If \emph a, \emph b, \emph c are the sides of any triangle, we know that \emph a+\emph b--\emph c will always be a positive quantity, because that is implied in the definition of a triangle.%%%%% \footnoteB{As a mathematician I would say that the assertion is the content of Proposition 20 of Book I Euclid's \emph{Elements} \cite{MR17:814b,Euclid-homepage,MR1932864}, which is implied ultimately by Euclid's postulates and ``common notions''; Moderns may say there are some ``hidden assumptions'' involved in the implication as well. In any case, as far as Collingwood can tell (or I, for that matter), only logical implication can give us infinitely many conclusions.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Thus the functional theory presupposes the rationalistic or implicational theory, which I have already given reasons for rejecting. (\gr b) I do not know whether Earl Russell, in the sentence quoted above, wished to be understood as meaning that \uline{the word `necessary'} has no other meaning than that which he there ascribes to it. If so, he was mistaken. It \uline{has another meaning, which is in fact its original meaning}. Just as the original sense of the word `cause' is what I have called the historical sense, according to which that which is caused is the act of a conscious and responsible agent, so the original sense of the word `necessary' is an historical sense, \uline{according to which it is necessary for a person to act in a certain way}: deciding so to act and acting therefore freely and responsibly, yet (in a sense which in no wise derogates from his responsibility) `necessitated' to act in that way by certain `causes', in sense I of the word `cause'. Even if Earl Russell does not wish to deny that the word `necessary' has this historical sense, I cannot think that his failure to mention it is well advised. This original sense of the word `necessary' is just as much the foundation on which the other senses of the [\textbf{321}] word `necessary' have been built, as the corresponding sense of the word `cause' (sense I, the `historical' sense) is the foundation on which have been built the other senses of the word `cause'. Between the respective histories of these two words there is not only parallelism, there is interconnexion. It is therefore very natural that Earl Russell should appeal to the word `necessary' in his attempt to clear up the meaning of the word `cause'. But the metaphysical problems connected with the idea of causation are historical problems, not to be solved except by historical treatment; and if the history of the word `necessary' has run on parallel lines to the history of the word `cause', the appeal from the latter to the former is scientifically barren, because it takes us not from one problem to the solution of that problem, but from one unsolved problem to another unsolved problem of the same kind. Most%%%%% \footnoteB{This paragraph could begin with ``(iv),'' as giving Collingwood's own answer to the question of ``what is meant by saying that a cause `produces' or `necessitates' its effect'' in sense III.} %%%%%%%%%%% people%%%%% \label{most} %%%%%%%%%%% think that when we use the word `causation' in sense III we mean to express by it something different from logical implication, and something more than uniformity of conjunction, whether observed only, or observed in the past and also expected in the future; and that this `something different' and `something more' is in the nature of compulsion. On the historical issue of what has actually been meant when words have actually been used, this is correct. Earl Russell (op.\ cit., p.\ 190) argues that people cannot mean this because (as he very truly says) `where desire does not come in, there can be no [\textbf{322}] question of compulsion'. All the same, as I shall now try to show, they do mean this.%%%%% \footnoteB{They \emph{do} mean it, and yet they \emph{cannot} mean it, because it requires conflicting presuppositions. We are in fact capable of having conflicting presuppositions, as when, in Book VI of the \emph{Iliad} \cite{Iliad-Loeb}, in meeting with Andromache and their son Astyanax in Troy, Hector is both sure that the city will be destroyed, and confident that the boy will have the chance to rule the city.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \uline{Causation in sense III is an anthropomorphic idea}. Natural scientists have tried to use it as a weapon for attacking anthropomorphic conceptions of nature; but it has been a treacherous weapon. It has led them unawares to reaffirm the view they were attacking. And \uline{that may be why, in Earl Russell's own words, `physics has ceased to look for causes'} (op.\ cit., p.\ 180). We found the idea of compulsion present in sense II of the word `cause'.%%%%% \footnoteB{On page \pageref{compulsion}.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% From what impression, we then asked, is this idea derived? We now find it present in sense III, and we must ask the same question, and answer it in the same way. The idea of compulsion, as applied to events in nature, is derived from our experience of occasions on which we have compelled others to act in certain ways by placing them in situations (or calling their attention to the fact that they are in situations) of such a kind that only by so acting can they realize the intentions we know or rightly assume them to entertain: and conversely, occasions in which we have ourselves been thus compelled. \uline{Compulsion is an idea derived from our social experience, and applied} in what is called a `metaphorical' way \uline{not only to our relations with things in nature} (sense II of the word `cause') \uline{but also to the relations which these things have among themselves} (sense III). Causal propositions in sense III are descriptions of relations between natural events in anthropomorphic terms. The reason why we are in the habit\label{habit-3} of using these [\textbf{323}] anthropomorphic terms is, of course, that they are traditional. Inquiry into the history of the tradition shows that it grew up in connexion with the same animistic theory of nature to which I referred in discussing sense II of the word `cause', but that in this case \uline{the predominant factor was a theology of Neoplatonic inspiration}. If a man can be said to cause certain events in nature by adopting certain means to bringing them about, and if God is conceived semi-anthropomorphically%%%%% \footnoteA{I distinguish an anthropomorphic conception of God (cf.\ p.\ 185) from a semi-anthropomorphic. An anthropomorphic God would be simply what Matthew Arnold called a `magnified nonnatural man'. His attributes would be merely the attributes of man, enlarged. For example, he would be liable to anger, but his anger would be a more formidable thing than man's. A semi-anthropomorphic God would be the result of criticizing this childish idea in the light of the reflection that, if God is really greater than man, he cannot have those attributes which in man are due to man's littleness; e.g.\ anger, which comes of being thwarted.%% \footnoteB{The sense of ``semi-anthropomorphic'' will be spelled out further on page \pageref{semi-anth}; meanwhile, Chapter X, ``Passion,'' of the \emph{New Leviathan} \cite[pp.\ 70 f.]{Collingwood-NL} spells out the meaning of anger: \begin{quotation} \textbf{10.\ 43.} In anger you have no consciousness of being angry; that comes only with reflection upon anger; what you are aware of is simply a contrast betweeen yourself and something (you know not what) other than yourself. This is the intellectual element in anger. It is identical with the intellectual element in fear. \textbf{10.\ 44.} The difference is purely practical. You conceive yourself as `contradicted' or `contrasted with' by the not-self (10.\ 27). The simplest thing to do is lie down under this menace. That is fear. The alternative is to rebel against it. That is anger. \end{quotation}} %%%%%%%%%%%%%%%% } %%%%%%%%%%%%%% as having faculties like those of the human mind but greatly magnified, it will follow that God also will be regarded as bringing about certain things in nature by the adoption of certain means. Now \uline{comes a step in the argument which, if we tried to re\-% construct it without historical knowledge, we should probably reconstruct wrongly}. If \emph x is a thing in nature produced by God as a means of producing \emph y, we might fancy \emph x to be a purely passive instrument in God's hand, having no power of its own, but `inert', as Berkeley in the true spirit of post-Galilean physics insists that matter must be. And in [\textbf{324}] that case God alone would possess that compulsive force which is expressed by the word `cause'; that word would not be given as a name to \emph x, and God would be the sole cause.%%%%% \footnoteB{Passive things cannot be causes. This is in accord with the assertion of Chapter XXX (starting on page \pageref{quod2}) that causes are not \emph{mere} situations or desires.} %%%%%%%%%%%%%%%%%%%%%%%%%%% Actually, \uline{God is for medieval thinkers not the sole cause but the first cause}. This does not mean the first term in a series of efficient causes (a barbarous misinterpretation of the phrase), but a cause of a peculiar kind, as distinct from `secondary causes'. The \emph{Liber de Causis}, a Neoplatonic Arabic work of the ninth century,%%%%% \footnoteB{According to \emph{Wikipedia,} this work was once attributed to Aristotle.} whose influence on medieval cosmology was at this point decisive, lays it down that \uline{God} in creating certain instruments for the realization of certain ends \uline{confers upon these instruments a power in certain ways like his own}, though inferior to it. Thus endowed with a kind of minor and derivative godhead, these instruments accordingly acquire the character of causes, and constitute that division of nature which, according to John the Scot, `both is created and creates'. Their causality is thus a special kind of causality existing wholly within nature, whereby one thing in nature produces or necessitates another thing in nature. The words `produces' and `necessitates' are here used literally and deliberately to convey a sense of volition and compulsion; for the anthropomorphic account of natural things is taken as literally true; the activity of these secondary causes is a scaled-down version of God's and God's is a scaled-up version of man's. This idea of God is only semi-anthropomorphic,\label{semi-anth} [\textbf{325}] because it implies the ascription to God of a power not belonging to man, the power of creating instruments of His will which are themselves possessed of will. This was the atmosphere in which our modern conception of nature took shape. For in the sixteenth and seventeenth centuries, when the animistic conception of nature was replaced among scientists and philosophers by a mechanistic one, the word `cause' was not a novelty; it was a long-established term, and its meaning was rooted in these Neoplatonic notions. Thus when we come to Newton, and read (e.g.)\ the \emph{Scholium} appended to his Definitions, we find him using as a matter of course a whole vocabulary which, taken literally, ascribes to `causes' in nature a kind of power which properly belongs to one human being inducing another to act as he wishes him to act. Causes are said, in the twelfth paragraph of that \emph{Scholium}, to be `forces%%% \label{scholium} %%%%%%%%%%%% impressed upon bodies for the generation of motion. True motion is neither generated nor altered, but by some force impressed upon the body moved.' The cause, for Newton, is not that which impresses the force, it is the force itself.%% \footnoteB{The quotation is from the \emph{Principia} \cite[p.\ 412]{MR1965137}. I wonder whether Collingwood work through the mathematical arguments of this work.} %%%%%%%%%%%%%%%%% Here and throughout his treatment of the subject it is perfectly clear that for him the idea of causation is the idea of force, compulsion, constraint, exercised by something powerful over another thing which if not under this constraint would behave differently; this hypothetical different behaviour being called by contrast `free' behaviour. This constraint of one [\textbf{326}] thing in nature by another is the secondary causation of medieval cosmology.%%%%% \footnoteB{\label{laws}% What exactly does Collingwood mean by Newton's ``treatment of the subject''? We shall come back to this. Newton's Second Law of Motion \cite[p.\ 416]{MR1965137} is, \begin{quote} A change in motion is proportional to the motive force impressed and takes place along the straight line in which that force is impressed. \end{quote} Proposition 7 (Problem 2) of Section 2 of Book 1 of the \emph{Principia} is, \begin{quote} Let a body revolve in the circumference of a circle; it is required to find the law of the centripetal force tending toward any given point. \end{quote} Thus some forces, at least, are to be accounted for by laws.} %%%%%%%%%%%%%%%%%%%% Taken \emph{au pied de la lettre,}%%%%% \footnoteB{Taken \emph{literally,} that is, and not merely ``properly'' in the sense of page \pageref{proper}, although the French definition of the phrase is ``\emph{Au sens propre, exact du terme}'' \cite[Lettre, p.\ 1477]{Robert}.} %%%%%%%%%%%%%%%%%%%%% Newton is implying that a billiard-ball struck by another and set in motion would have liked to be left in peace; it is reluctant to move, and this reluctance, which is called inertia, has to be overcome by an effort on the part of the ball that strikes it. This effort costs the striker something, namely part of its own momentum, which it pays over to the sluggard ball as an inducement to move. I am not suggesting that this reduction of physics to social psychology is the doctrine Newton set out to teach; all I say is that he expounded it, no doubt as a metaphor beneath which the truths of physics are concealed.%%%%% \footnoteB{Newton is \emph{not} expounding Neoplatonism, but is using its terminology, metaphorically. Likewise, Apollonius \cite[p.\ 6]{Apollonius-Heiberg} metaphorically calls a certain figure \gr{anaqwrht'hs}, ``one who retires from the world,'' which is ultimately from \gr{>an'a} and \gr{q~wros} ``space'' \cite{Skeat}. In particular, the digraph CH in ``anchoress'' is etymologically correct. It is \emph{incorrect} in ``anchor,'' which ultimately derives from \gr{>'ankura}, which does mean anchor, but is also the Greek name of the city of Ankara.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% and the quotation is from Part V (of eight), ``Of Schrifte // \emph{Of Confession}'', which begins \cite[p.\ 298]{Ancren-riwle}: \begin{quote} Of two \th inges nime\dh\ \.geme,%%%%% \footnoteD{Following the practice of \cite{OE}, I use ``\.g'' in place of the ``insular g'' (ancestor of the letter yogh) in the text.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% of schrifte, i\dh e beginnunge. \TH et forme \th ing, of hwuche mihte hit beo. \TH et o\dh er \th ing, hwuch hit schulle beon. \TH is beo\dh\ nu ase two limes; and ei\dh er is to-dealed; \th e uorme%%%%% \footnoteD{The editor and translator observes in his introduction, ``\emph U and \emph v are used indiscriminately for each other, and for \emph f.''} %%%%%%%%%%%%%%%%%%%%%%%% o six stucchenes;%%%%% \footnoteD{I use a semicolon for what in the text looks like a colon whose \emph{upper} dot has been given a rising tail.} %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \th e o\dh er o sixtene. // \emph{Concerning confession. To begin, take notice of two things: first, of what efficacy it is; secondly, of what kind it should be. These are two branches; and each of them is divided: the former into six parts; the other into sixteen.} \end{quote} The second of the two branches begins as follows, the enumeration being added by me \cite[p.\ 302]{Ancren-riwle}: \begin{quote} Loke we nu \.georneliche hwuch schrift schule beon \th et bere\dh\ swuch strenc\dh e; \emph{\&}%%%%% \footnoteD{The symbol with the meaning of an ampersand in the text is in form an italic \emph t with a curl flowing left, and then downwards and back, from its upper tip, as if to suggest a preceding \emph e.} %%%%%%%%%%%%%%%%%%%%%%% for to scheawen hit bet dele we nu \th is lim o sixtene stucchenes. \P\ Schrift schal beon \begin{inparaenum}[(1)] \item wreiful, \item bitter, mid seoruwe, \item ihol, \item naked, \item ofte imaked, \item hihful, \item edmod, \item scheomeful, \item dredful, \emph{\&} \item hopeful, \item wis, \item so\dh\ \emph{\&} \item willes; \item owune \emph{\&} \item studeuest; \item bi\dh ouht biuoren longe. \end{inparaenum} // \emph{Let us now consider attentively what sort of confession that must be which produceth such good effects; and to shew it better, divide we now this part into sixteen particulars. \P\ Confession shall be \begin{inparaenum}[(1)] \item accusatory, \item bitter and sorrowful, \item full, \item candid, \item frequent, \item speedy, \item humble, \item with shame, \item anxious, \item hopeful, \item prudent, \item true, \item voluntary, \item spontaneous, \item steadfast, and \item premeditated. \end{inparaenum}} \end{quote} The sixteen particulars are considered in turn, with examples. Concerning candidness, or ``nakedness,'' we are told, \begin{quote} Me Sire, \th eo wummon sei\dh, Ich habbe iheued leofmon; o\dh er, Ich habbe ibeon fol of me suluen. \TH is nis nout naked schrift. Ne biclute \th u hi nowiht. Do awei \th e totages, \th et beo\dh\ \th e circumstances. Vnwrih \th e \emph{\&} seie, Sire, Godes ore \emph{\&} tin! Ich am a ful stod mere; a stinckinde hore. // \emph{``Sir,'' saith the woman, ``I have had a lover;'' or, ``I have been foolish concerning myself.'' This is not plain confession. Put no cloak over it. Take away the accessories, that is, the circumstances. Uncover thyself and say, ``Sir, the mercy of God, and thine! I am a foul stud mare: a stinking whore.''} \end{quote} On the contrary, calling oneself a ``stud mare'' or ``stinking whore'' is not plain confession, but hyperbole. In any case, a bit later, still on the subject of nakedness, we have the following (again the enumeration is mine): \begin{quote}\sloppy Abuten sunne ligge\dh\ six \th inges \th et hit helie\dh; o Latin, circumstances; on Englisch, heo muwen beon ihoten totagges: \begin{inparaenum}[(1)] \item persone, \item stude, \item time, \item manere, \item tale, \item cause. \end{inparaenum} // \emph{There lieth about sin six things which conceal it; in Latin, circumstances; in English, they may be called adjuncts: \begin{inparaenum}[(1)] \item person, \item place, \item time, \item manner, \item number, \item cause. \end{inparaenum}} \end{quote} Concerning the first adjunct, we are told something of what distinguishes women from men (and here I do not bother to type up the original, only the translation): \begin{quote} \emph{Person---she that committed the sin, or with whom it was committed. Lay it open, and say, ``Sir, I am a woman, and ought rightly to have been more modest than to speak as I have spoken, or to do as I have done; and therefore my sin is greater than if a man had done it, for it became me worse\lips Sir, it was with such a man;'' and then name him---``A monk, a priest, or clerk, and of such an order, a married man, an innocent creature, a woman, as I am.''} \end{quote} So it continues through the other adjuncts. At the sixth, we have the \emph{OED} quotation with ``cause,'' and a sample confession about cause: \begin{quote} \emph{``Sir, I did it for pleasure, and for guilty love, and for gain, through fear, through flattery. Sir, I did it for evil, though no evil came of it. Sir, my light answer, or my light behaviour enticed him toward me. Sir, of this word came another; of this action, anger and evil words. Sir, the reason why the evil still continues is this: my heart was so weak.''} \end{quote} Here then is the teaching that women can \emph{cause} men to abuse them (and themselves). Two questions to be considered are: \begin{inparaenum}[(1)] \item Is this teaching true? \item Can the spread of this teaching among men \emph{itself} cause men to abuse women? \end{inparaenum} \chapter{Cause in Aristotle}\label{app:Aristo} \lofoot{Cause in Aristotle} Collingwood introduces \emph{causa ut} and \emph{causa quod} on page \pageref{4causes}. He calls them final and efficient cause, respectively, but we might spell out their Latin names as ``cause for the sake of which'' and ``cause that.'' They are two of Aristotle's causes, in the \emph{Physics,} Book \gr B, Chapter 3, 194\textsuperscript b 23--35 \cite{Aristotle-Physics-OCT,Aristo-Phy-Apost}: \begin{quote} \begin{compactenum}[(1)] \item \gr{t`o >ex o<~u g'igneta'i ti >enup'arqontos} that from which, as a constituent, something is generated; \item \gr{t`o e>~idos ka`i t`o par'adeigma} the form or the pattern; \item \gr{arq`h t~hs metabol~hs `h t~hs >hrem'hsews} that from which change or coming to rest first begins; \item \gr{t`o t'elos; to~uto d' >est`in t`o o<~u <'eneka} the end, and this is the final cause [that for the sake of which]. \end{compactenum} \end{quote} These are abbreviated as \begin{inparaenum}[(1)] \item material, \item formal, \item efficient, and \item final cause. \end{inparaenum} In note \ref{F=ma}, page \pageref{F=ma}, I consider Newton's Law of Gravity as a \emph{formal} cause. In Collingwood's Latin, the Latin adverb or conjunction \emph{ut} or \emph{ut\={\i}} is an adverb or conjunction with various translations into English, the relevant one at present being ``in order that'' %According to the \emph{Pocket Oxford Latin Dictionary} \cite{POLD}. In addition to being the neuter relative pronoun, \emph{quod} is a conjunction, with translations like ``because.'' Thus it corresponds to the Greek \gr{<'oti} and the English ``that,'' as is noted in the \emph{OED} article on ``that'' \emph{qua} conjunction. An illustration is from \emph{Isaiah} 63:5, in the King James Version of 1611: \begin{quote}\centering I wondered that there was none to uphold. \end{quote} I note how the Oxford World Classics edition of the KJV takes up Collingwood's theme of history in its preface \cite{KJV-Oxford}: \begin{quotation} \noindent\textsc{Bibles} are, by their very nature, partisan. As that plural suggests, there are many bibles, even in English, and each is the product of a particular interest group---% whether religious, commercial, or, increasingly nowadays, both. This edition is no exception\lips \lips Though this edition, unlike almost every other on the market, is not sponsered by a particular religious group, the mere fact that we have chosen to use a translation that is, in places, more than four hundred years old, indicates an initial historical bias to our approach. Bias is not confined to the choice of text\lips \end{quotation} \chapter{Motion in Newton}\label{app:motion} \lofoot{Motion in Newton} This is a remark on Collingwood's comment on page \pageref{inertia} at note \ref{inertia}, ``in Newton there is no law of universal causation.'' Here is Newton's First Law of Motion \cite[p.\ 416]{MR1965137}: \begin{quotation} \noindent\emph{Every body perseveres in its state of being at rest or of moving uniformly straight forward, except insofar as it is compelled to change its state by forces impressed.} Projectiles persevere in their motions, except insofar as they are retarded by the resistance of the air and are impelled downward by the force of gravity. A spinning hoop, which has parts that by their cohesion continually draw one another back from rectilinear motions, does not cease to rotate, except insofar as it is retarded by the air. And larger bodies---planets and comets---preserve for a longer time both their progressive and their circular motions, which take place in spaces having less resistance. \end{quotation} Thus motion need have no \emph{external} cause. And yet motion may continue because of a body's own ``perseverence'': might this motion not be considered as ``self-causing'' in the sense of Chapter XXX, page \pageref{self-cause}? Perhaps not. Newton's Definition 3 is, \begin{quotation} \noindent\emph{Inherent force of matter is the power of resisting by which every body, so far as it is able, perseveres in its state either of resting or of moving uniformly straight forward.} \lips inherent force may also be called by the very significant name of force of inertia. Moreover, a body exerts this force only during a change of its state, caused by another force impressed on it, and this exercise of force is, depending on the viewpoint, both resistance and impetus\lips \end{quotation} Here a ``state'' may be a state of rest or of uniform motion. So no exertion of force---no \emph{cause}---is required in either case. See page \pageref{vis-inertiae} and note \ref{inertia2}. Collingwood says he has ``already reminded the reader that in Newton there is no law of universal causation.'' Probably he is referring to his Chapter VI, ``Metaphysics as an Historical Science,'' where he says, \begin{quotation} I will go back to the example of causation, and remind the reader of three familiar facts. (\emph a) In Newtonian physics it is presupposed that some events (in the physical world; a qualification which hereinafter the reader will please understand when required) have causes and others not. Events not due to the operation of causes are supposed to be due to the operation of laws. Thus if a body moves freely along a straight line \emph p\textsubscript1, \emph p\textsubscript2, \emph p\textsubscript3, \emph p\textsubscript4\lips its passing the point \emph p\textsubscript3 at a certain time, calculable in advance from previous observation of its velocity, is an event which is not according to Newton the effect of any cause whatever. It is an event which takes place not owing to a cause, but according to a law. But if it had changed its direction at \emph p\textsubscript3, having collided there with another body, that change of direction would have been an event taking place owing to the action of a cause\lips \end{quotation} Collingwood will make similar remarks on page \pageref{ppp}. Again, the body's continued motion is indeed by a law, the First Law. A change in direction is due to the action of an external ``cause,'' that is, a force; but the action is still in accordance with a law, namely the Second Law of Motion, quoted in note \ref{laws}, page \pageref{laws}. It is not clear how much (if at all) Newton differed from the Moderns, by Collingwood's account, still in Chapter VI: \begin{quotation} (\emph c) In modern physics the notion of cause has disappeared. Nothing happens owing to causes; everything happens according to laws. Cases of impact, for example, are no longer regarded as cases in which the Laws of Motion are rendered inoperative by interference with one body on the part of another; they are regarded as cases of `free' motion (that is, motion not interfered with) under peculiar geometrical conditions, a line of some kind being substituted for the straight line of Newton's First Law. \end{quotation} Kepler showed that the planets moved along the lines called ellipses, which (if circles be counted among them) are themselves one of the three kinds of conic sections. These had been discovered by the Greeks two millenia before, perhaps by Menaechmus, probably for the sake of finding two mean proportionals between two given lengths. Newton shows that the motion of the planets along ellipses can be accounted for by an ``inverse square'' law of force. Collingwood makes it sound as if Newton was interested only in straight lines. Thus I wonder whether Collingwood has worked through Newton's mathematics. \chapter{Rule in Kant}\label{app:Kant} \lofoot{Rule in Kant} This continues note \ref{Kant-rule}, page \pageref{Kant-rule}. The Guyer--Wood edition \cite[p.\ 757]{Kant} of Kant provides a glossary of ``philosophically significant terms,'' but ``rule'' is not one of these: presumably it translates \emph{regel} \cite{GE}, which would seem, like ``rule,'' to be a descendent of the Latin \emph{regula.} The original meaning of this is one meaning of our ``ruler'': a straightedge. One can then imagine a sequence of events, lined up as if by a ruler, each being the cause of which the next is the effect. The sequence is not in space, but in time. The passage cited by Collingwood does indeed talk about sequences. It is in \begin{quote} \begin{compactitem}[--] \item \textbf{I. Transcendental doctrine of elements} \item \textbf{Second Part.} Transcendental logic \item \textbf{Division one.} Transcendental analytic \item Book II. Analytic of principles \item Chapter II. System of all principles of pure understanding \item Section III. Systematic representation of all synthetic principles of pure understanding \item 3. Analogies of experience \item B. Second analogy: principle of temporal succession according to the law of causality. \end{compactitem} \end{quote} At the head of the passage in the B edition is the assertion, typographically centered \cite[p.\ 304]{Kant}, \begin{quote}\centering All alterations occur in accordance with the law of the connection of cause and effect. \end{quote} This is followed by ``\textbf{Proof,}'' also centered. Collingwood refers to the fifth paragraph after this in the B edition; the paragraph is, \begin{quotation} In our case I must therefore derive the \textbf{subjective sequence} of apprehension from the \textbf{objective sequence} of appearances, for otherwise, the former would be entirely undetermined and no appearance would be distinguished from any other. The former alone proves nothing about the connection of the manifold in the object, because it is entirely arbitrary. This connection must therefore consist in the order of the manifold of appearance in accordance with which the apprehension of one thing (that which happens) follows that of the other (which precedes) \textbf{in accordance with a rule.} Only thereby can I be justified in saying of the appearance itself, and not merely of my apprehension, that a sequence is to be encountered in it, which is to say as much as that I cannot arrange the apprehension otherwise than exactly in this sequence. \end{quotation} Briefly, it seems, when we see a connection, it must be according to rule. Collingwood takes the connection for a ``one-one relation.'' Newton's First Law of Motion (note \ref{inertia}, page \pageref{inertia}) is about motion in a straight line, as if along a ruler; but it is not about a ``sequence of appearances'': \begin{compactenum} \item It is about not \emph{appearances,} but an idealized situation that never actually happens. \item It is not about a \emph{sequence,} for the points of a straight line do not compose a sequence. There is no one point that \emph{follows} a given point on a line. According to the glossary mentioned above, ``sequence'' is Kant's \emph{Folge,} which seems to be cognate with our ``follow.'' \end{compactenum} \bibliographystyle{plain} \bibliography{../../../references} %\printindex \end{document}