When something is tested to destruction, for instance, when a seat is repeatedly pounded by a weight attached to a robot arm to gauge how long it will last in regular use, a series from the past traverses the present and extends into the future at each thump on the cushion.
According to Deleuze’s philosophy of time, past series are concentrated into the present blow, which is itself ‘contemplating’ them. Contemplation is passive absorption and change; for example, when a split on a cushion grows wider as the weight hits. There is also active change in the present, when something new appears, like a branching in the split.
For Deleuze, as a way of taking up other times, the present is composed of three processes: contemplation of the past; active change through the new; and ‘expectation’ as momentum into the future. The tear in the cushion comes from the past, becomes new and different, then projects into later series.
I was asked recently, through this website, whether I could give a simple explanation of this philosophy for a reader with limited philosophical background but interested in Deleuze’s ideas. In addition to my reply, I’d suggest thinking of his work as a practical observation of the effects and significance of repetition over time in a technical situation.
These technical experiences are familiar to us, not only because we depend upon repeated interactions with mechanical and biological extensions, but because they matter. We feel this when things fail: a worn knee joint; a machine part beginning to clank and grind; lassitude and tiredness when stuck in a mental rut, each day the same but harder than the last. We also feel it in the desire for things to repeat: delight in a familiar mechanism; in a repeated gesture; in an event brought back to us.
Prompted to give an accessible explanation of time and repetition, I recalled the testing of Boeing 777 wings. They were pushed to destruction through the incremental addition of force to both wings. Bending dramatically, the wings withstood far beyond the threshold for certification: testimony to the strength of the 777 airframe.
As Deleuze points out in Difference and Repetition, although a series might be repetition of the same, there are also necessary differences; such as the increasing force on the wings. When the curved aluminium composites shattered under tension, the past series of increments was contracted into an explosion, followed by many changes projecting into the future, in the shards, shockwaves and enthused engineers shown in the linked video.
The wing is contemplating the build-up of force just before and during the fracture – as if the wings were like children, contemplating the increase of heat in their palms, when they dare each other to hold them over a candle. The earlier increments are concentrated into series leading to the failure of metal composites or nerves. This concentration is also an anticipation of future differences; in how close to failure we think we can now push, or in greater resistance built into wing design, or in calloused flesh.
Though I have used the candle analogy to convey physical and mental build-up then release, contemplation is not a metaphor. Deleuze’s terminology is directly descriptive rather than allusive. He means the wings are absorbing forces and changing passively in relation to them. There’s a combination of accumulation over time and across series with the beginnings of subsequent effects. These later effects are described as belonging to contemplation, because the series collect in the present while anticipating future changes.
Contemplation isn’t special or rare. Every present is a contemplation of earlier series, holding them together before they break into new movements. This ‘every’ is the beginning of an answer to an objection, not only to Deleuze but to philosophy more widely. If a philosophical description is not a metaphor or an analogy, how can it depart from accounts of the same processes given by contemporary physics, since the science offers the most accurate accounts of material processes?
A philosophical discussion is designed to be broader and further-reaching than science, not only in terms of features and types of process, but also in terms of past theories, contemporary clashes of ideas and the potential for future divergences. It’s a metaphysics over time – a comprehensive but necessarily speculative description of the world – rather than current science.
Philosophy is not simply free to depart from science. It has different roles and responsibilities. Philosophy learns from contemporary sciences and mathematics, but also from their historical evolution. Its conjectures are speculative, yet also anchored in lessons from the past, from a maximum of sources. For instance, philosophy can draw on artistic and literary forms of sensibility, meaning, observation, creativity, practice and imagination. Perhaps it should always do so, if it is to include feeling and imagination in its understanding of events.
The combination of speculation and a wide set of anchors or testing grounds means that the scientific standard of falsifiability applies only partly to philosophy. Networks of interconnected speculative proposals and multiple applicable circumstances can provide basic propositions ready for empirical falsification, but this falsification applies to propositions taken independently, with no equivalence to their function in the full network. There has to be a good reason for philosophy to depart from or raise questions of science, but this reason does not have the tightly defined criteria that make a scientific hypothesis falsifiable. This does not mean philosophies cannot be shown to be wrong.
Arguments and tests for philosophical disagreements often depend on critical hypotheses and counterfactuals (if so/what if), on the examination of words and concepts, on points about logic and clashes between different logics, on paradigmatic cases, examples and thought experiments, on contrasts between different philosophical schools and figures, and on historical cautions (yet it happened), limit cases (who is excluded) and varieties of experience (it can be felt differently).
Even if they are not directly under the rule of any scientific method, philosophy and history have reasoned arguments applicable to the sciences. Science is then restrained by criticism around imagined futures and past experiences. It is reminded of the risks of unexpected outcomes and the traps and entailments of both precision and imprecision of words and concepts. It is faced by logical fallacies and by the weaknesses of different logics.
Historical and philosophical research can make science aware of the dangers of paradigms and generalities insensitive to variation and change. They can draw its attention to speculative examples and to thought experiments. They recall the dangerous power of movements and figureheads. They warn against the inevitable corruptions of politicisation coming with the social benefits of science. They teach the hard lesson of how the best intentions have often led to the worst outcomes. Above all, they testify to the propensity of truths, beliefs and theories to come with unexpected, ignored, hidden or deliberately concealed costs to peoples, individuals, places, animals, plants and planets.
This critical restraint of science by philosophy and history is all the more important when the sciences become dominant yet often misunderstood standards for truth, value and practice. It is needed when science is strongly connected to economic returns and rewards; when it is organised in ways incentivising success and punishing failure; when it encourages treacherous emotions of ambition, greed and fear.
Philosophical arguments can adopt tests and experiments taken from modern sciences, but they also rely on different approaches distant from scientific practice. These arguments refer to longer time periods: epochs with tangled relations, such as historical revision of past glories into failures, or present caution based on imagined futures, or the haunting of the present by a past poorly understood and learnt.
Deleuze refers to contemplation partly in response to philosophical and mathematical problems. Contemplation connects series over time against the idea that time is a collection of independent instants. This independence leads to the paradox of separate moments that nonetheless belong together (causally, for example). Contemplation is also a response to the idea that if time can be divided infinitely then we can conceive of it progressing but never reaching a later point (versions of Zeno’s paradoxes of motion). Counter to discrete moments and infinite divisibility, contemplated time is constituted by continuities over series – durations – such that you cannot separate a present contemplation from its past and future without loss.
The Boeing experiment invites questions about duration, instants and series because it is a duration, since the test involves increases over time, and a series of instants, since after each incremental increase there is a pause and the wings are monitored for stresses at that time. In the pause, a state of the wings is registered (curvature, forces at different points, integrity of different materials and stability of the structure). Some of these increments are given more significance than others; the threshold for certification, for instance.
There’s an argument from physics and mathematics against the idea that we have a series of static instants indicated by specific measurements. At those instants, there is still change at molecular and sub-molecular levels; for example, in the progress of a fracture as understood through molecular dynamics. Metal alloys aren’t static and the philosophical case could incorporate this dynamism.
Independent of the argument from physics, there is also a philosophical reasoning for duration and continuity in series. The point is about context, significance and value. Instants are brought together in durations defined by multiple and continuous contexts, types of significance, and ranges and degrees of changing values. If we observe the Boeing experiment, each instant of the test has a wider context of how the test is set-up, where the instant is situated in the series, what properties are being monitored, how they are recorded, why, where, at what date, who by and for whom.
Instants also have particular significance, as shown by dramatic ones where the metals delaminate. A typical Deleuzian remark is that significance applies to all instants, not just some. For example, there is significance in how far an instant is from apparently more important ones (because importance changes over time). After an accident, features of the incident can lead to a review of significant thresholds and readings from earlier testing – say when a rotor blade fails in flight at an unexpected rate of revolution, leading to a review of the effects of the combination of turbulence and metal fatigue at speeds thought to be well within safe operating guidelines.
I have highlighted Deleuze’s insistence on ‘every’, ‘all’ and ubiquitous significance (nothing is inconsequential). They are important ontological, ethical and political features of his philosophy, illustrating a critical gap between science and philosophy. Science focuses on significant points such as maxima and minima, regularities, clusters and inflection points. Philosophy recognises this significance but maintains critical distance because, when different epochs and perspectives are taken into account, other points come to the fore, due to changes in context, shifts in ideas of significance and altering values. For instance, over recent years, the importance of outright performance of systems has given way to greater emphasis on their efficiency, due to the new context of climate change, new significance of energy use and novel values of care for endangered environments.
As an aside, ubiquitous significance is the reason commentators on Deleuze’s work on differential calculus and on Leibniz might have misunderstood his metaphysical independence from calculus. Every point on a curve is significant, not some, because significance is a property of relations to other curves and contexts (mathematical and other). That’s also why Deleuze adds Meinong and the Stoics to Leibniz: multiple and sometimes incompossible orders of significance subsist together and every grain is significant in many ways, not just some.
Changing contexts of value have effects upon relations of instants; not only monetary values, but also moral, aesthetic and scientific ones. The increments after the certification point are not only evidence for the strength of the design but also of costly redundancy. In addition to engineering constraints on expenditure, redundancy has a moral point in acceptable risk and how we assign financial value to lives. In practice, this morality is often far-removed from any claimed universality of values and some lives have been less valued than others. Is this money well spent?
Problematic aesthetic and scientific values are harder to detect. They can be found, for example, in the attachment of engineers to particular kinds of experiment or in the admiration of enthusiasts for iconic designs. The key to discovering these values is desire: deep commitment to an approach, or powerful attraction to an object beyond its functionality.
It didn’t make pure engineering sense to keep Concorde flying so long, since much more efficient designs were available, but pilots, engineers, operators and passengers were attracted to the romance – an unstable mix of imagination, desire and sensuality – of the plane as well as its speed.
In recent years, scientific values attributed to virtual simulations as opposed to full-scale tests have shifted, such that the gold standard of how to test has moved in the direction of virtual modelling, not only for financial reasons, but because computers, small-scale complex processes and large-scale human interactions need to be tested together – for instance, in numerical simulations for pedestrian-induced vibrations on footbridges.
The lesson is that even the type of tests has a durational and continuous context underlying a few dominant models. Standards shift according to a complex environment, because types are situated within wider debates, practices, theories, laws and technological changes affecting what constitutes a good and effective experiment.
The instruments monitoring the wings and the physical theories the machines rely upon are part of an empirical experiment leading to conclusions about the sturdiness of wings. It would be foolish to ignore those findings and the practical and theoretical set-ups that permit them. Philosophy should be as consistent with them as it can, yet also provide a critical counterbalance connecting to wider problems.
There is no limit to these problems. They might include questions about the gap between experimental practice and everyday use. They might extend to lessons learnt from physics and engineering for different fields. Given recent history, they likely should consider the wider pressures and influences – monetary, ethical, aesthetic, social, historical and political – surrounding the design, testing, manufacture, marketing, sales, operation, servicing, environmental impact and global use of planes.
The extent of problems, their urgency and their relation to engineering in its many forms came into sharp and tragic focus around the role of Boeing’s Maneuvering Characteristics Augmentation System (MCAS) in crashes of 737 MAX planes in 2018 and 2019. There are narrow engineering aspects to the accidents, notably around the effect of component failures on computing systems (they set off multiple false alarms) and the need for redundancy in critical systems (a single component failure contributed to bringing the flights down, with no cross-checking against other data inputs). However, the National Transportation Safety Board’s recommendations go well beyond the design of MCAS.
The recommendations include:
- consideration of the effects of alerts on pilots’ ‘recognition and response’ (pilots were startled and overloaded by multiple warnings unforeseen by Boeing)
- consideration of pilot procedures and training, including manufacturers’ assumptions about how pilots might respond to emergencies (during the initial tests of the MAX, conclusions were drawn from flights by test pilots warned about how MCAS would take over and pitch the plane down; these conclusions were applied to less experienced pilots deprived of that foreknowledge)
- advice for national certification bodies to strengthen their procedures (certification of the 737 MAX had been given over to the manufacturer much more than for earlier designs, removing the protection afforded by external scrutiny independent of the manufacturer’s financial, time and organisational pressures).
These points show the human factors surrounding systems, such as training, information flows, workloads and cockpit resource management (how crews interact). They draw out political factors, like the will to decrease red tape, shrink the role of the state, protect national champions and respond to powerful lobbies. They also imply financial and therefore time factors, including the pressure to bring the MAX into production quickly and as cheaply as possible in response to competition from Airbus.
There were direct mechanical causes of the crashes – sensors broke. There were direct computing causes – the physical failure caused the planes’ computers to command repeated nose down pitch that the pilots tried to and eventually could no longer counteract. However, the problem presented by the crash is also economic, social, political and philosophical in ways going far beyond physics and engineering, but that should (and sometimes do) inform design and engineering.
It is important to resist a temptation at this point. One of the legacies of arguments about the scope of philosophy, art and science is to divide them into proper domains or cultures, where each holds sway without the others (science-knowledge; art-beauty; philosophy-wisdom; economics-finance; politics-government). This is a mistake because the problems they encounter go across domains and arise because of interference between them (for instance between politics, economics, practical wisdom and knowledge in the case of certification of the 737 MAX).
Flying highly engineered machines – among complex systems responding to weather, traffic and scheduling – the crew of a plane is constituted by economic, political and sociological movements and trends, in addition to technical understanding of flying skills. The problem of whether and how a pilot reacts to dangerous situations is not simply a matter of individual ability.
The following influences can weigh on the performance of pilots: how far cost is a factor in training pilots (the cost of simulator time or type training), in organising their work schedules (tiredness, concentration) and in keeping their skills up-to-date (constant updates from manufacturers and airlines); who gets to become a pilot and why (conversion from the military, the movements of pilots around the globe according to market pressures); how pilots are funded and promoted (debt, seniority); what moral values are put to the fore (independence, teamwork, obedience); the aesthetics of communication and ergonomics (ambiguous messages in times of stress, Aviation English as an imperfect subset interacting with other languages, positioning of controls and warning lights, distracting warning sounds); which psychological strengths are emphasised and which weaknesses rejected (single-mindedness, hesitancy); the role of private companies, the military and national schools in training; the ongoing costs and profitability of airlines and their long-term relations to manufacturers (cost-cutting, contractual pressures).
Even the solution of designing pilotless aircraft will have human elements – in the fear of flyers, love of flying and need for rapid travel over long distances; in calculations of acceptable risks, costs and losses; in human contributions to the design and to its aims; in the role of humans in manufacturing and distribution; in how humans react when planes still have accidents.
A technical problem is at the heart of an entanglement of many different series and times. These can range from the evolution of a specific component, to misuse by an unexpected customer base, to financial imperatives and ethical questions. In order to solve the problem, a strict engineering approach sets limits to these series through narrow definitions, questions and tests (If we can’t state the problem, we can’t even know where to begin in solving it).
I say strict because wider series are brought in – about use, cost, attractiveness, practicality, elegance and preferable solutions. Nonetheless, narrowness is perceived as essential to the scientific approach. It works in abstraction in order to progress safely on the basis of accurate definitions, verified theory, ongoing reviews and repeated testing. Though narrow abstraction is established practice, it isn’t necessary. Many problems have been solved by accident, followed by observation and skill.
There is nothing new to noting the relation of abstraction to wider contexts and different approaches to complex and evolving problems. Design methodologies have long taken account of this, notably in software design. The significant difference is rather that philosophy does not seek abstraction primarily to solve concrete challenges. Instead, the point is to offer models for the abstract form of problems.
By ‘abstract form’ I mean the relations that constitute the problem as a process shared by many different concrete cases without being fully determined by them. An abstract form traces the underlying problematic relations for many actual and imagined cases, but it cannot be reduced to them. The model for these forms is a diagram or plan of how different areas of the problem are connected to others through ongoing transformations. This isn’t a scientific theory explaining different phenomena, but a much wider speculative explanation. It might incorporate elements of theory, yet goes beyond them in connecting very different areas involved in mutual transformations: an abstract picture of how worlds change connectedly.
A concrete solution is meant to overcome a practical difficulty. An abstract philosophical model explains how problems come about, work and persist. The two are related, since a concrete difficulty is a sign of a philosophical problem, and since the model for such a problem could guide concrete solutions. This guidance can be explicit; for instance, when philosophy directs action. It is more often implicit and indirect, when philosophy has been absorbed into ways of thinking and acting – idealism, rationalism, empiricism or pragmatism.
Deleuze’s teacher and supervisor of his thesis on Spinoza, Ferdinand Alquié, explains the gap between concrete objects and what philosophies model (in Descartes, Kant and his own philosophy of surrealism) through an ontological background that doesn’t coincide with the object. Philosophies look for signs of patterns beyond phenomena, but since these signs appear as phenomena there are connections between two realms that nonetheless diverge.
When philosophy becomes a social movement – like rationalism – the gap between ontology and concrete objectivity is frequently forgotten, to the detriment of philosophy. Against the glib label of rationalism, Alquié shows how Descartes can be read positively as unreasonable and surreal: ‘[Descartes] est le plus irréationnaliste des philosophes‘. The problematic connection across an ineliminable gap dividing abstract form from apparently more concrete evidence explains why philosophy has a critical relation to science. Philosophy conforms to scientific advances, yet has independent metaphysical and ontological resources to foreground and review them.
Similar to opposition to philosophy as a pointless departure from physics, there could be an objection to philosophical work on the form of problems, as a futile distraction from the practical resolution of concrete challenges on the basis of established science and well-regulated experiments. Taken as an historical claim this point doesn’t hold up well, given close and fruitful exchanges between philosophy and more concrete subjects; for instance, in Descartes and Leibniz, and more recently through logicians such as Boole, Frege and Ramsey.
Though the objection to a philosophy of problems is weak historically, it can be stated more strongly as a defence of science-based technical expertise, in an age where science and technology have proven records of success at resolving problems. Laurie Anderson’s irony about experts and their hold over problems gives voice to this argument while calling it into question.
I have outlined philosophical responses to the objection about futility. Philosophy offers a broader reach. It reviews the history of past theories and their evolution. It assesses contemporary clashes and considers future effects and divergences. Philosophy sets out critical hypotheses and counterfactuals. It specialises in the analysis of language and its deconstruction. It cautions about flawed logics, the risks of dominance by established schools of thought and the hidden influence of external powers such as finance, politics, emotions, sections of society, or social norms. Philosophy pays special attention to limit cases, exclusions and varieties. It spans longer periods, right into imagined futures. It charts the work of context, significance and value within other disciplines and their interconnections.
Without denying the knowledge of experts, it is possible to draw attention to risks in handing over the definition of problems to them. To be able to solve precise challenges, experts tailor problems to suit their fields and knowledge. Yet this solving of narrow problems can make connected ones worse. The evolution of problems – against a background of contexts, significance and values – therefore provides a critical perspective on problems and solutions. Varieties of perspectives lead to differences about solutions, expertise and the persistence of problems.
In car manufacturing, hybrid engines solve the challenge of producing more energy from less fossil fuel, according to tests, but in cars that overall consume more resources in real world situations. In designing a crypto currency, experts solved the problem of how to make the chain of transactions secure, yet contributed to extreme energy consumption much of it from fossil fuels. Societies strive for increases in life expectancy, remove incentives to have children and restrict immigration, in the name of wealth, cohesion and happiness. Nevertheless aging populations suffer from economic stagnation, social tensions and healthcare crises setting the initial problem in a wider context at odds with more narrow solutions. This is not to say that restricted solutions are wrong on their own terms. It’s to say that the special role of philosophy is to provide abstract, creative and critical models giving problems wider contexts, thorough investigations of significance and cautious examinations of claimed and implied values.
Deleuze’s abstract model for problems is based around a timed logic. He suggests a form for series mapped on how times take one another and themselves as dimensions: how a process occurring in one time has an effect on another. This might seem a superfluous conceptual innovation, but it is necessary for Deleuze’s model to explain processes over time. It can be understood from more familiar ways of thinking about time. When we plan for the future in the present, the future is taken as a dimension. When we are trapped by the past, the present is taken as a dimension of the past.
I have discussed the logic corresponding to the present taking the past as dimension for technical series such as incremental additions of pressure over time. This is the logic of contemplation: the present absorbs and changes with those additions. When the future is taken as dimension by the present there is a logic of anticipation: the future is given shape in the present. When the present takes itself as dimension there is a logic of active change in the present: something new is brought into existence in the present.
The abstraction of these forms is important and explains why they should be described as logics. They claim to hold for all series, but give minimal content to them: past series are concentrated into the present, as the present introduces change and sets off anticipations into the future. Concretely how they are concentrated, changed and anticipated is a matter for observation and experimentation. Abstractly that they must be concentrated is a formal requisite.
This distinction explains the critical and creative roles of philosophy, in contrast to the observations, theories, hypotheses and experiments of science. It is not for philosophy to say that something was not observed correctly according to scientific norms, that a theory is false, an hypothesis unpromising, or an experiment botched. It is for philosophy to say that an observation is narrow, a theory too simple, an hypothesis restrictive and an experiment too detached.
Philosophical criticism depends upon creativity, because new models and ideas provide new perspectives for critical points. These points aren’t only about restrictiveness and limits. They are about exclusions and omissions. They are about forgetting lessons from the past, overconfidence in the present and lack of imagination about the future. Since any present is a contemplation of the past, it is always an error to assume we are free of the past. The challenge for any problem is to detect how the present has absorbed the past, what changes have occurred in the present and how this has led to anticipations of the future.
To see the formalism of Deleuze’s timed logic at work we can also look at how the past takes other times as dimensions. When the past takes the present as dimension it makes the present pass. This gives a formal process of necessary passing with two features: passing away and entering the past. Every present must pass and every present becomes part of the past. Again there is no concrete content to these processes, but there are strong critical consequences and directions. It is an error to think that anything is eternal or outside time. It is a mistake to think some present durations don’t enter the past. We should reflect on how a present is made to pass and how it becomes a matter of past record.
When the past takes itself as dimension the formal process is for the whole of the past to be drawn together and transformed. This strange idea can be summed up as a claim that the past is always changing and that nothing from the past is independent of that change. It is therefore an error to think that we can discard parts of the past in relation to the present and future. It is a mistake to think that the past is fixed and a matter of record. On the contrary, the way the past works on the present and future is always changing and when we consider a current problem we should think about how it relates to a changing past and whether parts of the past have been wrongly excluded.
As for the future, when it is taken as a dimension by the past it becomes an unstable conjunction of destiny and freedom. The past turns the future into destiny: a set of limits and patterns that are difficult to escape and hard to bear, but defining of individuals and communities. Yet, destiny is not necessity or full determination. Individuals and communities are free to counteract their destinies, not to deny them, but to replay them differently. Though the future is hostage to the past, the nature of this dependency is to include freedom to shape it.
Deleuze’s timed logic is a complex abstract form describing dimensions of times. As such, it is also a set of claims about concrete and technical series. No technical series is independent from the past, all of the past. Every technical series is an anticipation of the future, in ways that make a destiny and leave it free to be played out differently. No series is forever; it is becoming obsolete and part of a past that must return differently in every new series. Every series is significant. The point of Deleuze’s philosophy of time is to help us understand how.