Some Contemporary Reflections on Bergson’s Time and Free Will

By relying on a quotation by F.L. Pogson1 to be found in the introduction to his translation of Bergson’s Time and Free Will essay, in this paper I discuss Bergson’s theory of the relation between time and free will by analysing three key notion of his philosophy, namely that of creative power of nature, of memory and of duration. In particular, I discuss a possible reinterpretation of his metaphysics by using the growing block model of reality, and his anti-associationist philosophy of mind—by him identified with determinism—which seems to require a finite duration of the present experience. Finally, I propose a reinterpretation of his conception of free action by using Kane’s contemporary approach to the problem.2 1 Some Methodological Considerations There is little doubt that Bergson’s philosophical work has always focused on the nature of time, in particular on the difference between the time of our direct experience and the time of physics.3 Consequently, Bergson’s critique of physical time—regarded by him as a mere spatializing abstraction from the real duration—was not only caused by the philosophical and scientific challenge emerging from Einstein’s two theories of Relativity, but predated their famous meeting on April 6, 1922.4 Some contemporary reflections on Time and Free Will will show  In this paper I will follow Pogson’s 1913 translation of the original text, as it has been checked by Bergson himself (Bergson 2001).  I thank two anonymous refereed for their important suggestions and Alessandra Campo for her decisive and patient editing help. This paper has benefitted of the grant progetti di ricerca di rilevante interesse nazionale financed by the Italian Ministry of Public Education—Bando 2017 Prot. 2017ZNWW7F.  An informative and accessible survey of different views on the nature of time in the 20 century is in Fano and Tassani (2002).  Of course, Canales to whom we owe the complete and fascinating reconstruction of their meeting is well aware of this fact. See Canales (2015). OpenAccess. © 2022 Mauro Dorato, published by De Gruyter. This work is licensed under the Creative Commons Attribution 4.0 International License. https://doi.org/10.1515/9783110753707-007 in what sense Bergson’s earlier criticism of physical time affected his later confrontation with Relativity. ‘Contemporary’ as it is used here implies a conceptual evaluation of Bergson’s arguments that is independent of a philological attention to the literature in psychology and neurophysiology that Bergson had read before the completion of his work in 1899. In the following, I will therefore first briefly expose his conception of the creative power of nature and of duration, and then compare his notion of the relationship between time and free will with some relevant contemporary scientific and philosophical theories. 2 Various Approaches to the Problem of Free Will and Bergson’s Position Schematizing, in the history of western philosophy we find three different ways of articulating the inextricable conceptual link between time and free will: (1). Aristotle’s logical discussion in the ninth book of De Interpretatione of the so-called “future contingents” (non-necessary future-tense statements) —was based on the thesis that if we attribute a definite truth value to such statements and at the same time accept bivalence, we are not free. If it is now true that tomorrow there will a sea battle, the captain is not free to decide whether he wants to attack or not, since he cannot but attack. The same conclusion holds if it is now false that tomorrow there will a sea battle. And since in virtue of bivalence, a statement is either true or false, in both cases we are not free. (2). The well-known theological problem of a omniscient or prescient God who, knowing in advance what each of us will do, necessarily deprives us of the power to choose freely. (3). The more serious scientific threat to our free will, coming from determinist laws conjoined to initial conditions These logical, theological and scientific arguments threaten in a similar way our free will, which typically requires as a necessary condition the possibility to choose between a set of different, real alternatives. Bergson’s approach to the problem and his defense of free will is targeted against the third conception of determinism and is based on the idea that, thanks to our direct intuition of the continuous flow of our inner experience, we are immediately aware of our freedom. Despite my ‘reconstructive’ approach, in order to summarize Bergson’s philosophical attitude toward the issue of free will, it is necessary to put it in the context of his general philosophical outlook. In order to go some way toward the achievement of this very ambitious aim in the very small space of a paper, I have decided to rely on a quotation from F.L. 66 Mauro Dorato


1S ome Methodological Considerations
There is little doubtthatBergson'sphilosophical work has always focused on the nature of time, in particularonthe difference between the time of our direct experience and the time of physics.³ Consequently, Bergson'sc ritique of physical time-regarded by him as am ere spatializing abstraction from the real duration-was not onlyc aused by the philosophical and scientificc hallengee merging from Einstein'stwo theories of Relativity,but predated their famous meeting on April 6, 1922.⁴ Some contemporary reflections on Time and Free Will will show  In this paperIwill follow Pogson's1913 translation of the original text,asithas been checked by Bergsonh imself (Bergson 2001).  It hank two anonymous refereed for their importants uggestions and Alessandra Campo for her decisive and patient editingh elp. This paper has benefitted of the grant progetti di ricerca di rilevante interessen azionale financed by the Italian Ministry of Public Education-Bando 2017 Prot.2 017ZNWW7F.  An informative and accessible survey of different views on the natureoftime in the 20 th century is in Fano and Tassani (2002).  Of course, Canales to whom we owet he completea nd fascinatingr econstructiono ft heir meetingi sw ell awareo ft his fact.S ee Canales (2015). in what sense Bergson'se arlier criticism of physical time affected his later confrontation with Relativity. 'Contemporary' as it is used here impliesaconceptual evaluation of Bergson'sarguments that is independent of aphilological attention to the literature in psychologya nd neurophysiology thatB ergson had read before the completion of his work in 1899.I nt he following, Iw ill thereforef irst brieflye xpose his conception of the creative power of nature and of duration, and then compareh is notion of the relationship between time and free will with some relevant contemporary scientific and philosophical theories.

2V arious Approaches to the Problem of Free Will and Bergson'sP osition
Schematizing,i nt he history of western philosophyw ef ind three different ways of articulating the inextricable conceptual link between time and freew ill: (1). Aristotle's logical discussion in the ninth book of De Interpretatione of the so-called "future contingents" (non-necessary future-tense statements) -was based on the thesis that if we attributeadefinitet ruth valuet os uch statements and at the same time accept bivalence, we aren ot free.I fi ti sn ow true that tomorrow therew ill as ea battle, the captain is not free to decide whether he wants to attack or not,s inceh ec annot but attack. The same conclusion holds if it is now false that tomorrow therew ill as ea battle. And since in virtue of bivalence,astatement is either true or false, in both cases we are not free.
(2).The well-known theological problem of aomniscient or prescient God who, knowingin advance what each of us will do, necessarilyd eprivesu so ft he powert oc hoose freely.
(3). The mores erious scientific threat to our free will, coming from deterministl awsc onjoined to initial conditions These logical, theological and scientific arguments threaten in as imilar wayour free will, which typicallyr equires as an ecessary condition the possibility to choose between as et of different,r eal alternatives. Bergson'sa pproach to the problem and his defense of free will is targeted against the third conception of determinism and is basedo nt he idea that, thanks to our direct intuition of the continuous flow of our inner experience, we are immediatelyaware of our freedom. Despite my 'reconstructive' approach, in order to summarize Bergson'sp hilosophical attitude towardt he issue of free will, it is necessary to put it in the context of his general philosophical outlook.
In order to go some waytowardthe achievement of this very ambitiousaim in the very small spaceofapaper,Ihave decided to rely on aquotation from F. L.
Pogson⁵ extracted from his introduction to Bergson's Time and Free Will essay, that he translated 1910. "If aman weretoinquireofNature the reason of her creative activity and if she werewilling to give ear and answer,she would say -'Ask me not,but understand in silence,evenasIam silent and am not want to speak" (Pogson VIII, in: Bergson 2001, my emphasis). My choice is plausiblyjustified by the fact thatBergson himself agreed with the translator to use it as amotto summarizinghis own philosophyupto1910,amotto that is an epigraph takenfrom Plotinus.
My attempt at this point is to try to expand on the keyw ords appearingi n the succinct quotation aboveb yd iscussingt hem in some more details.

Nature
First of all, accordingtoBergson, Nature (with acapital N) is active: as such, it is not to be conceivedasaCartesian clockwork composed by corpuscles impacting one another,orasNewtonian particlesattracted by forces. Bergson'sconception of nature is more similar to Leibniz's, even if the philosophical system constructed by the German philosopher is rather different.⁶ Despite his anti-cartesian conception of nature,Bergson inherited from Descartes aradical dualism between matter and mind, mirrored by his epistemic dualism between the twofaculties of intellect and intuition. The intellect is afaculty that is essential for the effectiveness of our action in the material world because it is capable to separate and distinguish the objectsi ns pace and to fix in time the continuous, unified and temporal flux intuited by our consciousness (see also Russell 1914, 6). It must be kept in mind that, accordingt oB ergson, it is this experiencedc ontinuity (of which we are immediatelycertain) thatp revents an application of the concept of causality to our mental states.

Creativity and Novelty
The other buzz word, creativity,cannot be discussed without making referenceto the notion of élanv ital as it is defined in the CreativeE volution, something that here, of course, cannot be done in full. In our context,s uffice it to sayt hat the creative,active power of Nature stands for the power to generate something new that did not exist before and that is unpredictable from the past.Inhis opinion, this kind of novelty is incompatible both with Darwin's "chance and necessity approach" and with ateleological account of living beings. In fact,Bergson'screative power does not just concern living beings, but penetrates the whole of Nature and thereforep ervades also inorganic matter.
With respect to this view,i ti si ndispensable to recall the proximityofBergson'sn otion of duration to Whitehead'sc oncept of process,whose main philosophical outlook was greatlya ppreciated by the French philosopher,⁷ given that the formerextended such ac oncept to all matter,animated and inanimate. Aprocess for Whitehead is something that cannot be crystallized as achangeof thingsintime but is the basis of the identification of the being of the world with its becoming. The moment of becomingisindeterminate, since it is onlyafter the processhas taken place that can we regarditasafact and therefore as subjectto causation.C learly, the perception of an event thati sh appeningi nt he present can be fullyd escribed onlya fter the event has taken place, but accordingt o Bergson it can be made sense of onlywhen the whole constellation of memories that is related to it emergest oo ur mind, or only post factum so to speak.
Three main remarks must be pointed out at this stage, in particular vis àvis the role of the past in Bergon'sn otion of creativity.
(1) The first point involves the psychic experience of ag rowingp ast and its ontological correlate. Givenhis idea that each present experience P occurringat at ime t 1 is pregnant with ag rowingp ast, P is different from anyo ther experience, past or future. The reason is that P also contains the immediate, continuouslyf lowing memory of all past experiences: our past experiencesa ffect the wayw el ivet he present ones. In addition, rememberingt he past,a sP rimo Levi alsoh ad it (1969), is not a passive but a creative act,given that the content of the present,unique recollection R of apast event E is constantlyinfluenced by whatever has been experienced after the direct experience of Ea nd before R.⁸  Foracomparison of Bergson'smetaphysics with Whitehead'sphilosophy, Iamgreatlyindebted to one of the two referees,whose suggestions have been extensively used in the following lines.  Acontemporary explanationinscientific terms of this essential phenomenon is giveninEdelman (1980). The samedestinyoccurs to R after its occurrence, so that our memory of E is always different.T he key differenceb etween spatial and temporal experience is well expressed by the following passage: Our experienceofduration is unlikethe quantitative homogeneity presupposed by the concept of space: things in spacecan be counted because they areexternal one to the other in ah omogenous medium. On the contrary,o ur immediatee xperience of time (duration) is heterogeneous because each immediatelygiven data is always different and yeti su nified by an act of 'unification of disparate content': 'several conscious states areorganized into a whole, permeateo ne another, [ and] graduallyg ainaricher content' (Bergson 2001,1 22). This constantlyc hangingi nterpenetration of different temporale xperience makes time whollyd ifferent from the multiplicityo fo bjectsi ns pacea nd is what Bergson calls duration;t he richerc ontent of our experience of time is due to the above-mentioned role of memory,e xtendingm ore or lessi nt he past (immediate, closer or more remote). Counting objectso nt he contrary,p resupposest ime (as in Kant) but the activity of countingi sp ossible onlyb ecause each counted object is external to anyo ther.
On the contrary,w ec annot "count" the different temporale xperiencesi n succession: Bergson's "organization in the whole of several conscious states" points to the well-known fact that asuccession of experience is not an experience of succession. The latter is asynthetic act.Inorder to gain adeeper understanding of what Bergson calls "unification of disparate content" one can refer to contemporary phenomenological discussions about the extendedness of the experienced present,t he so-called "Specious Present",aterm coined by Clay( 1882) and then further articulated by the Americanpsychologist and philosopher William James (1890) and manyo thers after him. Our experience of music is often used to illustrate conceptions that are similar to Bergson'sh olistic approach to temporalexperience (seefor example Husserl 1990): when we hear asuccession of af ew notes forming am elody, we unify in as ingle, simultaneous experience what is temporallys uccessive (the single notes). Contemporary discussions⁹ about the psychologyo fo ur temporale xperience are thereforeh ighlyi ndebted to Bergson'ss eminal intuitions.
Even though Bergson does not go so farastoimplythatthe psychicnovelty of the contents of our mental acts depends on what we would call todaya"Growing Block Model Of Reality",Idaresaythathewould not be, at least prima facie, whollyu nsympathetic with such am etaphysical underpinning of our subjective experience of accumulations of memories. The so-called "GrowingBlock View of Reality",originallyformulatedbyBroad (1923), claims that there is literallynothing after the present moment (the future so to speak is "empty"), while the present is the moment in which new elements of reality that did not previouslyexist become actual. What has become present remains foreverpart of reality,since it must be presupposed to explain the incessant creative power of Nature.¹⁰ There is an obviousr eason to associate the GrowingB lock metaphysics to Bergson'sv iew of time. In fact,i tw ould explain our experience of an always richer temporalexperience, givenbythe accumulation of memories with the constant changeoffutureevents becomingreal in the present.The constant increase of our memories corresponds to, and can explained by,acontinuous "accumulation" of real facts or events in the world. It is interesting to note that am etaphysical model accordingt ow hich, as is typicalo ur understandingo fo ur life, each passingd ay is closer to the moment of our death, and more and more remote to the moment of our birth, would contradict Bergson'sv iew,s ince in this "erosionist" model, the past would become less and less real.¹¹ In his Introduction to metaphysics,these intuitions are expressed by Bergson with the metaphor of the twos pools "with at ape running between them, one spool unwinding the tape, the other winding it up" (see Lawlor and Moulard 2020).
(2)T he second remark anticipated at p. 68 is the fact thatB ergson rejected anyDarwinian explanation of the origin of new species: as such, his natural philosophyruns counter to what today(and to alesser extent alsoathis times) is a very well-confirmed scientific theory. To the extent that anaturalistic metaphysics requires that metaphysical hypotheses should at least be logicallyc onsistent with the resultso fo ur best scientifict heories, Bergson'sm etaphysical view should be rejected, based as it is on an anthropomorphic view of life, according to which "life, from its origin, it is the continuation of one and the sameimpetus, divided into divergent lines of evolution" (Bergson 1911, 61). Interestingly,his major objection against Darwin'sidea of natural selection is frequentlyv oiced also todayb yc ontemporary creationists: it is impossible to explain the complexity and the teleological features of living beingsbyintroducing what he calls a "mechanistic hypothesis".The mechanistic hypothesis thatheis referringtocorresponds to Darwin'snatural selection acting on variations dueto contingency and chance: "What likelihood is there that,bytwo entirelydifferent  Iowe the idea to explain Bergson'sthesis of novelty with this metaphysical view to Simone Gozzano, with whom Id iscussed it at some length.  See Casati and Torrengo( 2011);Norton( 2015). series of accidentsbeing added together, two entirelydifferent evolutions will arrive at similar results?" (Bergson, 1911, 63 -64,m yemphasis). As in Duration and Simultaneity-vis àvis the twin paradoxand the special theory of Relativity-also in this case his criticism to Darwinism is based on as erious misunderstanding, accordingt ow hich "twod ivergent branches of evolution converget ot he same point" (Bergson, 1911, 63 -64).
This is not what happens in the evolution of species as it was described by Darwin and as we know it today: the divergence of traits does not implyconvergence as Bergson has it,but rather more divergence. Supposing that Bergson had read Darwin'smain book (he quotes from it), in his understanding of Darwinism he neglects completelythe fact thatnatural selection, acting differentiallyonthe constantlychanging DNAoft he various species, favort hoset hat are capable of reproduction by giving birth to more descendants. In Bergson'si nstead, the creative evolution of life works against inert matter by promotingc omplexity.
In aw ord, accordingt oB ergson, Darwin'sm echanistic explanation is not sufficienttoexplain the origin of novelty in nature, which depends on avital impetus (élan vital)t hat overcomes the resistanceo fi nert matter,ahypothesis which to ac ontemporary reader remains completelyu nexplained.
However,a ccording to Bergson, neither Darwinian mechanicism nor am ere teleological approach to evolution would ensure the kind of novelty thath ei s after,s ince also within the latter approach the whole (the telos)w ould already be givena tt he beginning,a si nafoetus.
The doctrineofteleology,inits extreme form, as we find it in Leibniz for example,implies that thingsa nd beingsm erelyr ealize apreviouslyarranged program. However,according to Bergson, "if there is nothingunforeseen, no invention or creation in the universe, time is useless again. As in the mechanistic hypothesis, here again it is supposed that all is in advance. Finalism, thus understood is onlyi nverted mechanism" (Bergson 1911, 45,m ye mphasis).
From this passage, it is clear that Bergson identified the "creative power" of nature producing novelties with the unforeseen or the contingent,which takes us to our third remark.
(3) The third wayo fu nderstanding the notion of creativity comes from the opposition between determinism and indeterminism. In this third sense,which is the most important for the purpose of this paper,accordingtoBergson novelty requires the failure of determinism (see the sentencei talicized above). Parentetically, Bergson had receiveda ne xcellent mathematical education, and he was certainlya ware of the locus classical of the definition of determinism givenb y Laplacei nh is 1820's Treatise on Probability: We ought to regardthe present stateofthe universe as the effect of its antecedent stateand as the cause of the statet hat is to follow.A ni ntelligencek nowinga ll the forcesa ctingi n naturea tagiveni nstant,aswellast he momentary positions of all things in the universe, would be ablet oc omprehend in one single formula the motions of the largest bodies as wellasthe lightest atoms in the world, providedthat its intellect were sufficientlypowerful to subject all data to analysis; to it nothingw ould be uncertain, the future as wella st he past would be present to its eyes (quotedi nN agel 1961, 282-283).¹² It is often but incorrectlyalleged that determinism is incompatible with true novelty in the universe: in this case, we are told with amisleading metaphor that the whole temporale volution of the world would be "contained" in the initial conditions.¹³ But note that this would be true onlyi fd eterminism amounted to the epistemic notion of predictability as in Laplace'spassageaboveendorsed also by Bergson (seenote 12). But this is not the case, as deterministic chaos abundantly shows. Within chaoticphenomena, the evolution of asystem is dictated by a deterministic equation, but its evolution soon becomes unpredictable, wheret he "soon" in question depends on the system one is considering.¹⁴ It is onlyi fp redictability impliedacompleteinformation about the initial state of the universe (an impossible ideal) thatcould we claim with some reason that (1) in deterministic models of the universe the present and the futurea re all (epistemically/informationally) "contained" in the past so that (2)n oc reative power or novelty can be attributed to nature.
In other words, unpredictability is not sufficient to defend freew ill, since also within the GrowingB lock-in which future events don'te xist unrestricted-ly¹⁵-we have both completenovelty and at the same time adeterministic though thoroughly unpredictable evolution of physical systems. On the other hand, the block view of the universe, which is closed to novelties regarded as comingi nto being, is compatible with indeterminism. Since my third interpretation of Bergson'sview of the creative power of Nature is grounded in the text,wemust conclude that one cannot defend the view that novelty is guaranteed by unpredictability.A nd yet, manyf ollowers of Bergson, after the quantum revolutions,  Here is Bergson'sv ery similar reinstatement of Laplace'sp assage: "As uperhuman intellect could calculate, for anymoment of time, the position of anypoint of the systeminspace. And as thereisnothingmoreinthe form of the whole than the arrangement of its parts,the future forms of the systema re theoreticallyv isible in its present configuration" (Bergson 1911,11).  As we have seen, mutatism utandis for Bergson the same claim holds for at eleological account of life.  See Ruelle (1991).  See Mozersky (2011) for the meaningofunrestricted quantification; for instance, the current pandemic, in year 2018 was not existingand yetthe universe could be in principle deterministic. argued incorrectlythat the new Quantum Mechanics, allegedlyintroducing indeterminism in nature¹⁶,w as ac onfirmation of Bergson'sp hilosophyo fn ature.
One of the foundingf athers of the theory,L ouis De Broglie argued that "if Bergson could have studied quantum theory in detail … he could doubtless have repeated, as in TheC reativeM ind,t hat 'time is this very Hesitation or it is nothing'" (De Broglie 1941, quoted in Canales 2015. De Broglie clearly identifiesthe hesitation of the quantum system before measurement with the unreducible chance that the experimental apparatus reveals when the state of the system is not in ae igenstateo ft he observable. Keepingi nm ind that causal knowledge corresponds in the languageo ft he first twod ecades of the last century to Kausalgesetz or determinism, Canales tells us that the (alleged) attacks on causal knowledge due to Quantum Mechanics wasr egarded as consonant to Bergsonism by authoritative journals of philosophy. There is even aconversation that has been reportedaccordingtowhich Bergson would have said "that which is funniest [drôle], is thatp hysicists have come back to find liberty!" (Mondor 1957,6 9).
The illegitimate and superficial identification of Heisenberg'sprinciple of indeterminacy relations with af ailureo fc ausation (determinism) depends on the fact that the formulation of determinism needsa tt he samet ime ad efinite and simultaneous specification of position and momentum of aparticle, afact that in Quantum Mechanics cannot be achieved since position and momentum are conjugatev ariables and cannot be measured by the same apparatus at the same time. But this principle does not entail that the theory is indeterministic; in fact,w idelyd iscussed and exact deterministic formulations of QuantumM echanics (like that offered by Bohmian mechanics) assume that the quantum probabilities are epistemic.¹⁷ Furthermore, from am ore rigorous philosophical perspective,i ndeterminism, if regarded as imposing ab ranching structure to time-in which aunique past diverges into manypossiblefutures, all compatible with the samep ast-cannot be regarded as as ufficient condition for af ree choice in as in libertarian philosophical frameworks.C onsequently, the role of chance in anyhuman deliberation must be discussed less superficially: deciding what we oughtt od ob yt hrowinga" quantum coin" cannot the paradigm of a rational choice.A nd for the compatibilists,i ndeterminism is not as olution to the problem of the freedom of the will: one needs additional arguments to reject  Caution is needed because alternative formulations of quantum theory,l ikeB ohmian mechanics,a re fullyd eterministic. See Goldstein (2017).  See Goldstein (2017). the compatibilist view,accordingtowhich our will is free whenitisunconstrained, even if determined by previous desires.
In aword, if (a) the creatingp ower of nature in Bergson'ss ense consists in the fact that its temporal evolution produces unpredictable elements and if (b) freedom consists in this unpredictability,this third sense of novelty is as implausible as the second, which engages ad esperate fight against Darwin'se xplanation of the evolution of species. Forthis reason, aplausible interpretation of the creative power of nature is given, rather,b yametaphysics of aG rowingB lock, which might explain whyB ergson identified novelty and freedom: each act of our consciousness is free because, even on determinism, it is different from the previous ones despite the similarity of the circumstances of choice,a nd reality can be regarded at least in part as broughtabout by our intentions and actions. Iwill now try to connect this conception of freedom to some keypassages in the Essai sur les données immédiates de la conscience.

3F reedom in Time and Free Will
In the third chapter of the book Time and Free Will Bergson analyses very carefullyt he problem of determinism, regarded bothf rom the physical and the psychological perspective.Inthe physical realm,the principle of conservation of energy,a cting on corpuscles subjects to the lawo fa ttraction and repulsion is assumedfor the sake of the argument (Bergson 2001, 145). But Bergson decidedly questions its appropriateness for apsychological kind of determinism, since this thesis depends on the possibilityo fc onsidering our psychical states as invariablyc orrelated to physiological states.
Before discussingt he positive side of his conception of free will, we must stress his opposition to the claim that the principle of the conservation of energy-which became central in physiochemical phenomena after the rapid development of thermal physics¹⁸-also applies to living beings. There might be aspecial kind of energy in the latter entities, due to the fact that, unlike what happens in inert matter, there cannot be ar eturn to an initial state (except in intervals ot time exceeding the ageo fthe universe, see also below). The inapplicability to living beingsofthe conservation of energy and therefore, on Bergson'sview,todeterminism as he sees it,¹⁹ depends on the irreversibility of the development of  Sufficet or emind the readero ft he twof amousp rinciples of thermodynamics stated by the German physicist Rudolf Klausius "The energy of the world remains constant.The entropy of the world tends to am aximum." (Klausius 1867, 400).  To ac ontemporary reader,t his identification is dubious to sayt he least.
living beings and in particularo fconscious beings. Therefore, it depends on the different role playedbytime in inert matter and in our consciousness: "inert matter does not seem to perdure or to take anyt race of past time" (Bergson 2011, 153). There is no real changeininertial matter,which seems to remain "in an eternal present" (Bergson 2011, 153),but in the case of conscious beingsthe duration or the extension of apast sensation in the present one impliesac ertain kind of gain: the past is always real for anystate of consciousness,because anypresent sensation is affected by all of our past experiencei nt he ways ketched above.
The connection of free will with time should now be clear.I nanutshell, Bergson equates determinism with the principle of conservation of energy,a nd this principle in its turn with the (controversial)²⁰ idea that physical time is fundamentallyr eversible (Albert 2010). Since there can never be ac onscious state that repeats itself (each is pregnant with ad ifferent past),the principle of conservation of energy,a nd therefored eterminism, accordingt oB ergson, does not apply to consciouss tates. Beforep roceeding, it should be pointed out that Bergson'sa rgument is rather weak. He is certainlyo nt argetw hen he considers the principle of conservation of energy to be one of the strongest arguments in favoro fp sychophysical parallelism or even (though he does not consider this possibility) of an identity theory of mind and brain. However,a lso in physics macroscopic time is irreversible, and the ubiquityo ft he growth of entropy must be explainedi nt erms of time symmetric mechanical laws and initial conditions(presumably of the universe), in such away that the entropy of ac losed system never decreases.The return to the initial state, guaranteed by Poincaré's recurrence theorem (Sklar1993), requires atime that is longer than the ageofthe universe! Practically,t here is no difference between physical and psychological time as fara si rreversibility is concerned.
Furthermore, thereare some physical processes, called non-Markovian, that do "keep trace" of the past,unlike Markovian process, which have no memory of the past since theirp robability depends onlyo nt he previous states.B ergson is right on target when he stresses that in living beingsthe past is much morei mportant in determining their present and future states,but also the Earth,unlike an electron, keeps traces of its remotep ast,a nd the differenceb etween conscious statesand 'inert matter' in this respect could be regarded as one of degree.
Bergson continues his defense of free will by pointing out that thereare two kinds of time -at heme thatw ill return later in his debate with Einstein in Duration and simultaneity (1965). One is the abstract time presupposed by mathe- Some processes regarding muons in radioactive decayshow afailure of temporal invariance (Horwich 1987). matics and physics,the other is the internal time of consciousness, which is pure duration (durée).²¹ One of the lasting contributions of Bergson'sp hilosophyl ies in his deep reflections on the conflict between these two notions of time:t his question is still at the center of the contemporary philosophical debates on the nature of time. Bergson argues that, since manyphilosophers toooftenintermingle and confuse these two kindso ft ime,t hey come to think that also conscious states are subjectt op hysical time. Thisi st he main reason whyt hey don'tr egard our actionsa sf ree: abstract time is presupposed by the studyo f inert matter,whered eterminism reigns with no limitations.
The fundamental argument thatBergson puts forth to defend our freewill is his justified attack on associationist philosophyofmind, where the self is regarded (on the wake of Hume and Mill) as am ere collection of ideas and emotions, each of which is separated and external one to the otheri nahomogenous medium. In our internal life, on the contrary, Bergson correctlypointsout that conscious states "penetrate and melt one another,each tinged with the colouringof all the others" (Bergson 2001,165). This implies that each of us loves and hates in different ways (Bergson 2001,1 65), according to our different character,s ince each of us has had ad ifferent set of experiences, each of which reflects holisticallyall the others we had in the past like amonad reflects the universe. According to Bergson, the fact that our selvesare singular and unrepeatable is another argument in favoro ft he fact that we possess af reew ill since, not onlyi st he development of the self unpredictable and novel,b ut it is also characterized by duration,i .e., the permanence and fusion of all the relevant,r emembered conscious states constitutingatemporale xperience.A nd once more, the link between time and free will emerges: abstract time goes togetherw ith associationism, which in its turn implies psychological determinism, while duration implies ad uration in the present of all past conscious stateso faperson. In Bergson's view,aradicallyh olistic conception of selfhood guarantees our free will since someone'spersonality is not givenbythe sum of her conscious state "but is present in each of them.A nd the outward manifestation of this inner state is what we call afree action, since the self alone willbethe author of it,and since it expresses the whole of the self" (Bergson 2001, 165 -66).
Areconstruction of this argument might go like this. If one acts upon as ingle detached desire or reason, the corresponding act will not express the whole personality of the doer,s ince on deterministic associationism, the self is frag- It is ironic, even if not significant fromaconceptual viewpoint to remark that in the Scholium to the Principia Mathematica Newton referredt oa bsolute time, flowingequably, as duratio (duration)! mented into adetached series of conscious states,each of which is the deterministic cause of the next,a nd the personality resultsf rom the sum of causesa nd effects, each of which detached from the other.Adeterministic action,i nB ergson'sview,would not express 'the whole of the self''sp ast-and thereforeone's entire personality-but,atany present moment of her life, onlythe content of the present experience, caused by immediatelyp revious ones.
In contemporary terms, and stretching somewhat my hermeneutic/reconstructive approach, some such actions might even be regarded as the deeds of an addictive person, awanton-who is incapable of postponing the satisfaction of her current desire (Frankfurt 1982)-and thereforea st he actions of ap erson who discounts the futureand does not act in virtue of an intention that integrates her whole life. The free actionsa re those in which the present state of mind takes into account the imagined future andt he remembered past of the self,i n such away that an attitude thathas been appropriately called temporalneutrality is not violated.²² Fort his reason, Bergson claims, if each conscious state is, thanks to its duration, inextricablye ntangled with all the others, the action will express the personality of the doer,h er global, holistic self, and will therefore, in avery plausible sense, be free. In aword, accordingtoBergson freedom is self-expression, and thereforebasically autonomy: the unconscious determination of our will should be compared to hypnosis, aphenomenon that later interested also Freud.
Actions done under hypnosis are paradigmatic examples of heteronomous determinations of our will, which,inBergson'sunfair renderingofdeterminism, exemplifyawillg uidedb yexternal forces. Ther endering is unfair because also in compatibilist conceptions of freew ill, determinism is compatible with the claim that our will is free because we can do what we want,wheret he "we" refers to our will when it is not determined by external factors likehypnoticstates, threats,and impediments of this kind. One could claim that also on compatibilism, determined motiveso fo ur actions are the expression of our personality,a claim that, pace Bergson, could be jeopardizedonlybythe introduction of some form of chance or indeterminism separatingt he past from the future.
It is extremelyi mportant to stress the fact thatB ergson correctlyr ejects the implausible idea that we enjoy an absolute form of freedom. In almost all actions of our life, we are guidedbyasortofmemory that todaywewould call procedural,the kind of automatic memory thati si nvolvedw hen we ride ab ike or wash our hands. In these actions, there is no expression of the self and an association- See Brink (2011). Some ContemporaryR eflections on Bergson's Time and free will ist/deterministic description of the relevant action is appropriate:B ergson correctlyl ikens them to "reflex acts" (Bergson 2001, 168).
In these actions, languagep lays as ocial role that,however,crystallizes our deepests elf. In most of our waking hours, the means-to-end actions, in which what he calls the intellect plays av ery important,i ndispensable practical role, the self does not emerge but remains at the surface.IfIcan interpret rather freely Bergson'sp oint (but possiblyn ot too unfaithfullyt oh is explicit intentions) the rare but free actions are, instead, thosei nw hich "we formo urselves",t op ut it with Kane (2005,130), who refers to these as self-forming actions. Similarly,B alaguer (2010) refers to "torn decisions" as those in which our free is particularly manifest.S ince these actionsf ullyr eflect our past selvesi naholistic way, they also shape the type of person that we will become in the future. In Bergson'slanguage, areallyfreeaction expresses the whole enduringself, its past,its present and its future, in aword, its duration. It is appropriate to summarize what Ij ust wrotew ith the following quotation, resemblingK ane'sc onception of af ree action. "It is at the great and solemn crisis, decisive of our reputation with others, and yetm orew ith ourselves, that we choose in defiance of what is conventionallycalled amotive,and this absence of anytangible reason is the more striking the deeper our freedom goes" (Kane 2005,130).
My brief synopsis of Bergson'su nderstanding of the relation between time and free will would not be completeifIomittedtomention that the essentialdifferenceb etween time and space that he defends also serves to explain his criticism to the principle of alternative possibility( VanI nwagen 1986), on which almost every contemporary discussion on free will is based. Notably, in Bergson's opinion the branching time structure (one of the possible interpretations of his idea of novelty,see above) is inappropriate to defend or attack indeterminism or determinism. Ab ranching,n on-linear structure of time is ag eometrical representation of ad ecision process that for Bergson essentiallyi nvolves time and not space, because it is based on the becomingo ft he self and of the totality of reality.
The defender of libertarian conceptions argues that we could havedone otherwise givent he samep ast because the path (symbolizingt he course of actionw ea re about to take as in Fig. 1) "is not alreadyt aken, therefore it mayt ake anydirec- Fig. 1: At the bifurcation point, the free decision "brings" the historyo ft he self to the "right". tion whatever" (Bergson 2001, 182)²³ but then he continues "it is not possiblet o speak of apath till the actionisperformed. But then it will have been traced out" (Bergson 2001, 182).Infact,hecontinues, "before the path was traced out there was no direction either possibleorimpossible, for the very simple reason that it could not have been aquestion of apath" (Bergson 2001, 182).This reasoning is grounded on the presupposition that the principle of alternative possibility is based on as patialization of time: in this case, the attempt to represent it with afork (a path, see Fig. 1above,wherethe bolded, thicker path is the one actually taken by the world, and the other path remains ap ossibility). But freedom is pure duration, becoming, so it cannot be crystallized by representing it with a spatial, symbolic graph. The idea here is that anysymbolicrepresentation involving space, the metaphor of the fork, is static, and then incapable of renderingthe changei no ur conscious states in which becomingc onsists.E ven "the moving now conception of time" (Skow 2015) would render onlym etaphoricallyt he idea of the seamless changeo ft he enduringself. On the contrary,the metaphysics of the GrowingBlock explains the accumulation of our memories and therefore, givenB ergson'sv iew of temporal experience the fact that to bring into being ap reviouslyn on existing path and thereforeanon existingr eality.
One can accept his conception of selfhood and at the samet ime point out that his reasoning in the quotedp assagei sw rong. The evaluation of the future contingent certainlylies at the point of bifurcation, but even after acertain path has been taken, it makessense to claim, counterfactually, that if the evolution of the world had been governed by indeterministic laws, the future could have been different without presupposinga ny changei nt he past state of the universe. In deterministic models, on the contrary,w ec ould have done differentlyi ft he past had been different,sothatthe differencebetween deterministic/linear models and indeterministic/branching models remains even if the future is "empty".

4B ergson'sC riticismo fL anguage as an Instrument to Grasp Becoming
In various passageoft ime and freew ill, Bergson points out that language, with its cataloguingf unction typical of juxtaposition in space of different things,h as an extremelyi mportant practical role but is incapable of grasping the property that each enduringconscious state has to reflect throughout time the whole personality of ahuman being.Werefer to loveand hatred as if two different human  Here Bergson reproducesa lmost exactlyt he libertarian conception of free will.
Some ContemporaryR eflections on Bergson's Time and free will beingshad the same feelings, which lead us to supposethat they are in the same mental state. Languagemust generalize via universals that,however,are not apt to grasp the uniqueness of an individual, which is anecessary and sufficient condition for her autonomy. Pure duration escapes the limitation of language. This is the main reason why, in the opening quotation of the paper (Pogson's), silence is recommended as the best wayt ou nderstand the creative power of nature, which manifests itself at the highest degree in our consciousness. It is intuition, our direct awareness of our inner states that leads us to decide one course of action rathert han another. The intellect'so nlyf unction is to calculate the consequences of our choice. However,t his produces another drawback of Bergson'sc onception of a free choice. He says that looking back at an important decision, we have the impression that we acted without ar eason (Bergson 2001, 170). He is led (if not forced) to defend this claim because he wants to leave room to our intuition and not just to the calculating, spatializing intellect.A nd, paradoxically, he claims thatt he more an action is without ar eason, the more is free: we wish to know whyw ehavemade up our mind, and we find that we have decided without anyr eason, and perhaps against every reason.B ut in certain cases this is best of reasons.For the action which has been performed does not then express some superficial idea, almost external to ourselves, distinct and easy to account for,but agrees with our most intimated feeling, thoughts and aspirations, with that conception of life which is equivalent to all our past experience (Bergson 2001,1 70).
The obviouso bjection is whyaconception of life that reflect all our past experiences should not be ac onscious reason motivatingu st oc hoose one path rather than another.I ft he process that leadsu st od ecide one path rather than another is hiddenfrom us, it cannot be based on feelings, because typically we are aware of the feelings or emotions that moveustoself-forming actions expressingo ur past.F urthermore, the opposition between feelingsa nd reason is not supported by the contemporary neurophysiologicalresearch:²⁴ without emotions or feelingsn od ecision could be reached.
However,inorder to be more charitable to Bergson'sview, which should not be saddled with anachronistic considerations comingfrom contemporary discoveries but onlyw ith his own conceptual unclarities-this was after all the methodological assumption of this paper (see the introduction)-and in order to go some waytowardjustifyinghis position on the role of reasons in self-forming decisions of our life, we should consider thath ei ntended reason in av ery limited  See Damasio (1994) among others. sense, as a momentary cause of our action, which is called into playbythe practical needsofour dailylife. With this caveat in mind, we can agree with Bergson's claim thatt he reallyi mportantd ecisions in our life are not due to momentary inclinations, but are expression of our whole selvesr egarded as enduring "entities".Noduration of the selves, no self-formingdecisions. However,itremains to be shown whyt his sort of decisions are free, in the sense that they are not an expression of some form of causal determinism encompassingacomplex series of previous experiences. The fact thatatemporallyholistic conception of the self allows us to claim thatthe decision is adeep expression of the values of the individual person is not sufficient to claim that the motivation to follow one path rather than another is not fixed by previous events.
Afinal wordonPogson'sreferenceto"our silence and Nature's silence" contained in the quotation above. Silence is necessary for the liberation of the self from the practical needsofour dailylife based on our intellect that classifies objects on which we must act.Behind languagethereisspace,the outer world, and thereforethe crystallizing,abstract time of physics.But the true nature of the self is not expressible in words, because if this werepossible, we would fragment our conscious states,t hereby eliminating their duration in each of our actions. The elimination of this duration entails, as we have seen, ad enial of our freedom. The cost to gain freedom is high: Bergson argues in favoro ff reedom with more or less plausible arguments that are however formulated in linguistic form: languagei ss omething we must getr id of in order to perceive the unity of the self with the creative,f reep ower of Nature.S o, in aw ay,t his claim reminds us of the earlyWittgenstein'sapproach to philosophy: in the case of Bergson, one must use languageinorder to show thatlanguageisnecessary but not sufficientf or af ree action.