Skip to content

Ullin T. Place (1924-2000)

The basic form of causation

The basic form of causation is a description of causation that applies to all cases of causation, so it is claimed. In essence, causation is causal interaction in which an object, animal or person does something that changes an outside system. The basic form can be recognised in causal statements of day to day language. In this post, the basic form of causation is applied to causation by omission and simultaneous causation.


Although causation means many things dependent on the situation in which the concept is applied, in my view, a core, what I call the basic form of causation, applies to all cases of causation. This means that if the basic form does not apply, it is not a case of causation. The basic form of causation consists of four elements: a causal origin, a causal process, an effect system and an effect or outcome. The causal origin is cited when asked who or what caused the effect.[1] When asked how the effect came about, a description of the causal process is given.[2] These two kinds of questions can be answered in many instances by one statement with the form: “X caused Y by q-ing” or “X q-ed Y”. X and Y refer to the causal origin and the effect, respectively, and the q-ing belong to a subclass of the transitive verbs: causatives that take objective complements (Aronson, 1969, 1971a). For example, in “Peter pushed the door open”, ‘open’ is an objective complement. As Aronson (1969, p. 548; 1971a, note 9) notices, an objective complement “completes the description of the action expressed in the verb and modifies or qualifies the direct object”. In our example, Peter is the causal origin of the causal process of Peter pushing the door, the effect system, which has the outcome of the door being open (changing from being closed to being open).

We can refer to a cause by naming or describing the causal origin, by naming or describing the causal process or by naming or describing both, as in the last example: Peter pushed the door. The causal origin and the causal process always go together. If only one of the two is mentioned as the cause, this does not mean that the other is missing or does not exist; it is only not mentioned. More generally, only when all four elements of the basic form exist or are the case there is causation. Causation understood in this way can also be called causal interaction. In the words of Byerly:

I suggest we say that a system A (as a localized object) interacts with a system B to cause a change in B if there exists a causal process originating from A that changes system B. (Byerly, 1990, pp. 548-549)

System A is a causal origin that interacts with effect system B. The change in system B is the effect of the causal process originated by system A.

The phrase ‘causal origin’ is related to the observation by Anscombe (1971) that

Causality consists in the derivativeness of an effect from its causes. This is the core, the common feature, of causality in its various kinds. Effects derive from, arise out of, come of, their causes. … Now analysis in terms of necessity or universality does not tell us of this derivedness; rather it forgets about it. For the necessity will be that of laws of nature; through it we shall be able to derive knowledge of the effect from knowledge of the cause, or vice versa, but that does not shew us the cause as the source of the effect. (Anscombe, 1971, p. 67 of the reprint in Sosa, 1975)

Instead of ‘causal origin’, we could have chosen the phrase ‘causal source’.

In the example of Peter pushing the door, the transitive verb ‘to push’ describes the causal process. Anscombe (1971, p. 68 of the reprint in Sosa, 1975) gives the following examples of causal, transitive verbs: “scrape, push, wet, carry, eat, burn, knock over, keep off, squash, make (e.g. noises, paper boats), hurt.” Cartwright (2017) gives another list of examples, all related to her morning toast:

pressing the lever on my toaster lowers the spring-loaded rack where my bread sits, lowering the rack closes the circuit, closing the circuit switches on the heating element, the temperature rise expands a metal strip, … the movement of the catch trips a lever, the lever releases the toast rack, the rack springs back, loaded with the bread that has been browned by the same heating element that expanded the metal strip. (Cartwright, 2017, p. 137)

An interesting class of causal verbs is the ergative verbs, which are both transitive and intransitive. When it is transitive, the object is the same as the subject when intransitive. Transitive: ‘Peter opened/closed the door’, ‘Peter broke the glass’. Intransitive: ‘The door opened/closed’, ‘The glass broke’. The transitive sentences entail: ‘Peter caused the door to open/close’ and ‘Peter caused the glass to break’. When used transitively, the ergative verb does not say what the subject did, only that the subject did something, and it describes what happened to the effect system. The ergative verb describes what happened to the effect system when used intransitively but without referencing the causal origin.

The causal origin and effect system are also known as the causal agent and causal patient. The causal agent is said by UTP (Ullin T. Place) “to act on the object or patient whenever a cause or set of causal factors produces an effect” (1973-1974, p. 5 of lecture 5). The causal agent’s action (the acting) is the causal process that produces the effect on the patient.[3] Byerly (1979) calls causal origins substantial causes[4], referencing Aristotelian ‘primary substances’. For UTP, a causal agent is also a substance, in the sense of Aristotle. A substantial cause or causal agent is cited in sentences in the passive voice as the object of the by-clause: the puncture was caused by the nail; the accident was caused by Jones (see Byerly, 1979, p. 64). Baron-Schmitt (2024) talks about things[5], but

[n]ot all things are Aristotelian substances, and I want to allow that these non-substances can cause. My left earlobe, my belt buckle, and perhaps even my shadow can cause, even if they are not substances.

And

“A causes B to do β by doing α” is the most straightforward form that thing causation can take. (Baron-Schmitt, 2024)

Translated to the basic form of causation, ‘A’ is the causal origin, ‘B’ is the effect system, ‘doing α’ is the causal process and ‘to do β’ is the effect.[6] The bomb (A) causes the building (B) to collapse (do β) by exploding (doing α). The first two examples of things not being (Aristotelian) substances, a left earlobe and a belt buckle, are parts of an encompassing whole. It seems odd to say that they can cause by doing something or acting on another thing or object that leads to an effect, apart from the whole of which they are part. UTP (1973-1974, lecture 5) stresses that a causal agent (and patient), being a substance, must have a logically independent existence. The earlobe and belt buckle depend for their existence on the substance they are part of. What can they do that is not already a doing of the encompassing substance? Take a pushpin that punctures the tyre of a bike. Is the pushpin or only the pin of the pushpin the causal origin of the puncturing of the tyre by which it became flat? It would go too far to analyse this case in all its details, but I suggest that the whole pushpin is the causal origin that punctured the tube, not only the pin. A shadow also does not logically exist independently of the body it is a shadow of. Although one can say that a shadow does things, this is in a causally powerless manner.[7] But, what is interesting here is that a causal process that leads to an effect is a doing by the causal origin. Without a doing by the causal origin, there will be no effect. This means that so-called omissions, doing nothing, cannot have causal effects because no causal process is involved. Later, more on omissions.

For UTP, “the occurrence or existence of an effect entails the existence of a substance or other logically independent existent which we may refer to as the object or patient which undergoes the change or is in the state in question” (Place, 1973-1974, p. 5 of lecture 5). Thus, a causal patient is, according to UTP, an Aristotelian substance analogous to a causal agent. In my view, an effect system can be causal patient, but when an artist (the causal origin or agent) transforms loose ingredients into a piece of art, the loose ingredients are the effect system, which is not an Aristotelian substance.[8] In many cases, the effect system will not be a collection of loose ingredients but dependable, stable systems that behave in an orderly and predictive manner (see also my post https://utplace.uk/dependable-systems-why-is-psychology-possible/). In Pearl’s interventionist approach, which will be discussed in a later post, stable and autonomous causal mechanisms (that underly the arrows in a causal diagram) play an important role. I will argue that these causal mechanisms are effect systems in the sense of the basic form of causation.

In physical cases, the causal origin is a source of some physical quantity, such as energy or momentum. This quantity flows from the causal origin to an effect system. This flow is the causal process, which leads to a change in the effect system, the effect or outcome (cf. Aronson, 1971a, 1971b; Byerly, 1979; Fair, 1979). The direction of this flow or transference can decide the causal origin or agent and the causal patient. Aronson (1971a, pp. 424-5) describes the case of placing ice cubes in a glass of water. We say that the ice cubes caused the water to cool. Physics teaches us that heat flows from the water to the ice and is converted to kinetic energy, which counteracts the forces that keep the ice in a solid state. So, it is more accurate to say that the water caused the ice to melt by giving up some of its internal energy; in this process, the water became cooler. But what is the answer to what caused the water to become cooler? My answer is that it is placing the ice in the water (bringing the ice in contact with the water); the causal agent is whatever can do this.

In the previous paragraph, we saw that the physics that applies to the case being judged can determine our causal judgment. In general, our causal judgments depend on our world knowledge. Thus, scientific theories can play a decisive role in applying the basic form of causation. This also applies to causation in non-physical domains. I think this implies that there always exists a physical connection between the causal origin and the effect system. However, the physical connection is less important in understanding causation in non-physical domains. In the human domain, communication and the exchange of information between causal origins and effect systems superimposed on physical processes are the essential elements in causation.

Causation by omission

It is claimed that the absence of what usually causes something to happen is the cause of the non-happening; this is known as causation by omission. McGrath (2005) suggests that causation by omission always has a normative component. Whether or not we accept the existence of causation by omission (or by absence), this is an interesting suggestion because I think that causation, in general, involves normativity in the sense that causation involves a change in what would be otherwise normal.[9] This is what McGrath says about normal:

It is normal for x to φ iff x is supposed to φ. People are supposed to keep their promises (it is normal for them to keep their promises); alarm clocks are supposed to ring at the set time (it is normal for them to ring at the set time); hearts are supposed to pump blood (it is normal for them to pump blood); the rain is supposed to come in April (it is normal for it to come in April); water is supposed to flow downhill (it is normal for it to flow downhill).
    … [W]hat is normal [is] imposed by certain standards … So, for example, there are certain artifactual standards governing what things like alarm clocks are supposed to do, and if an alarm clock does not ring at the set time then it violates these standards. … There are certain biological standards governing what organs like hearts are supposed to do, and if a heart does not pump blood then it violates these standards. … There are even physical standards governing the natural world. Laws of nature, and regularities in general, set standards for weather and for stuffs: rain that comes in May rather than April, or water that flows uphill, violates these standards. (McGrath, 2005, p. 138-139)

It is generally said that a cause (a causal origin plus a causal process) makes a difference, manifested in its effect. Effects are changes in the effect system that would not have happened without the cause. But the effects are changes to what? Here, we can answer that the effects are changes to what would have normally happened. Causes are, in a sense, interfering in the normal course of an effect system or preventing the normal course of an effect system. In classical physics, the law of inertia (Newton’s First Law: a body remains at rest or moving with constant velocity in a straight line unless an external force acts on it) is an example of a standard of normality. Aronson (1971a, p. 421) calls being at rest or in constant linear velocity natural changes which do not need external causes to explain them. Natural changes in a system are changes that are normal for that system. We find the same idea in the classic on causation in the law by Hart and Honoré:

The notion, that a cause is essentially something which interferes with or intervenes in the course of events which would normally take place, is central to the common-sense concept of cause, and is at least as essential as the notion of invariable or constant sequence so much stressed by Mill and Hume. (Hart & Honoré, 1959, p. 27)

 Normality is a relative notion, but it is not subjective. Take the much-discussed example of the plant that died because the person who would look after it did not water it. Not all plants will die if they are not watered. It is normal for plants to die under certain circumstances. For example, houseplants kept inside will die when not watered, but they can survive if kept outside with occasional rain. Also, succulents, inside and outside, will survive long without water. So, it depends on certain factors what normally happens to a plant when it is left to its own devices. And this is a matter of fact, independent of what we humans think about how things normally go. This is even true in the case of what humans normally do because of the norms they agreed on. The norms are objective facts that are normally followed up.

When you think that causation by omission is possible, you have to explain why the not-watering by the responsible gardener is the cause of the plant dying and why the not-watering by, say, the queen is not the cause of the plant’s death. Both do nothing, and according to our basic scheme of causation, there is, in neither case, a causal process connecting the causal origin with the effect, and thus, the death of the plant is not caused by the gardener doing nothing or the queen doing nothing. The dying of the plant is what normally happens when it is not watered. There was no cause to interfere or to prevent. That the gardener is responsible for the dying of the plant and is to blame is a different matter. It is evident that the queen is not responsible.

McGrath (2005, p. 134) introduces the notion of a would-be preventer, a type of cause that would have prevented the event (supposedly) due to the omission. Watering-by-the gardener is a would-be preventer of the plant’s death. Then she introduces normal-would-be preventers Co of e that

are such that, relative to some actual standard S, had e been prevented, it would have been normal for an event of type Co to have prevented e. (McGrath, 2005, p. 141)

Here, McGrath applies the notion of normality to causal processes and not to effect systems, as I did in the case of the gardener and the dead plant. The standard that might apply to the behaviour of the gardener is that it is normal to keep yourself to the promises made. Relative to that standard, the gardener could have prevented the plant from dying by keeping his promise, and this would have meant watering the plant, which he did not do, which was his omission. But I do not accept the next step that McGrath has to take: the step from the gardener could have been causally effective in preventing the plant from dying, to by doing nothing, the gardener caused the death of the plant. The only way the gardener could be causally effective in the circumstances was by watering the plant, but he did not. He was not causally effective; the plant followed the normal course in the circumstances and died. The gardener was responsible for not being causally effective. His omission was a case of responsibility or, better, of not taking responsibility; it was not a case of causation.

Simultaneous causation

Mackie (1979, p. 23) describes Von Wright’s example of a box with two buttons on top. Pushing the one (A) to go down also causes the other (B) to go down and vice versa. This effect is due to the construction of the box. For Mackie, this illustrates simultaneous causes and effects: the going down of A is the simultaneous cause of the going down of B. But in my opinion, the buttons are both part of an effect system; the parts are fixed together. Pushing either button A or B causes the buttons to go down simultaneously. The pushing is a causal process outside the effect system containing A and B.

Similar cases are where what Hempel (1965, p. 352) calls laws of coexistence apply. A law of coexistence relates quantities such that changing a quantity will simultaneously change the other related quantities. Hempel gives the example of a simple pendulum where the period in seconds is proportional to the square root of the length in centimetres. Other examples are the gas laws of Boyle and Charles and Ohm’s law. A pendulum, a fixed amount of gas or a conductor, is (part of) an effect system in which when an outside cause changes a quantity, the other quantities are changed simultaneously. However, the quantities of a law of coexistence are not alike. An external cause can change the pendulum’s length, which is also a change of its period. However, changing the pendulum’s length indirectly by changing the period is impossible. The reason is that length and period are different kinds of properties. Length is a categorical property, while the period is a manifestation of a dispositional property (see for this distinction, e.g., Armstrong et al., 1996). A dispositional property is only manifest under particular conditions. The period of the pendulum is not observable when at rest. Only after the pendulum is set in motion does the period of the pendulum become manifest. The disposition underlying the period is not made explicit in the pendulum law.

Woodward (2003; pp. 197-198) uses the example of the pendulum to illustrate the problem of explanatory asymmetry. The period of a pendulum can be explained by its length, but the reverse of explaining the length of a pendulum by its period is not counted as an explanation. For Woodward, the crux is that the length of the pendulum can be manipulated to change the period, but there is no possible manipulation of the period to change the length of the pendulum. But what is missing in his account is why this is the case. In my opinion, the missing element is the idea that two different kinds of properties are related to each other in the pendulum law, and this difference accounts for the asymmetry.

For Leuridan (2012), the pendulum is an example of simultaneous causation, or in his words, a synchronic causal relationship:

a simple pendulum involves synchronic causal relations. The length l of a simple pendulum is rightly regarded as a cause of that pendulum’s period T. By intervening to change l, one changes T (Woodward [2003], p. 197). Yet the change of period does not follow the change of length in time. (Leuridan, 2012, p. 420)

The assumption is that the change in the pendulum’s length and its period are two different events that are causally related.[10] However, there is just one change caused by the intervention. The length is changed, which is a change in the disposition to show a certain period in the right circumstances.

I am committing to what Armstrong (1968, p. 86) called a Realist view of dispositions in which dispositional properties are identified with their categorical bases. In Armstrong’s writings, this categorical basis is always a microstructure of the thing or organism to which the disposition is ascribed. But, in the case of the pendulum, the categorical basis is its length. The period of a pendulum is independent of its material constitution. So, the microstructure of the pendulum does not play a role in its behaviour, while the period of its movement depends on the pendulum’s length. Place (1996e, p. 114) mentioned the sharpness of a knife where a dispositional property “is a feature of the macrostructure rather than the microstructure of the object.” The length of a pendulum is thus another example.[11]

My thesis is that external causes can change the categorical properties of an effect system, and dispositional properties can only be changed indirectly by changing a categorical property to which it is lawfully related. Let us look at the other examples of laws of existence mentioned by Hempel. The gas laws only apply when the gas is constrained in a closed environment like a chamber, container or cylinder. This closed environment is the effect system. The volume of the contained gas is a categorical property. With a moveable piston, the volume can be changed by external causes. If the chamber is isolated so that the gas temperature remains constant, the gas pressure, a dispositional property, is inversely proportional to the volume manipulated by outside causes. For Ohm’s law to apply to an effect system, a conductor or a resistor must be integrated into an electric circuit. It is less clear whether the resistance of a conductor (measured in ohm) is a categorical property. However, the resistance directly depends on the length and the area of the cross-section of the conductor, both categorical properties and the material used in the conductor’s manufacturing. Like the length of a pendulum, the resistance of a conductor is independent of its functioning. A pendulum lying down, not displaying its period and a conductor not integrated into a circuit still have the same length or resistance as when they are in use. When integrated into a circuit, the resistance of a conductor determines the ratio of the voltage measured over the conductor and the ampere of the current through the conductor according to Ohm’s law. This ratio can be changed by replacing the conductor with a conductor with a different resistance. Changing the voltage or the current will not affect the resistance.


[1] ‘Causal origin’ is a more or less literal translation of the Dutch ‘oorzaak’ and the German ‘Ursache’. The ‘oor’ or ‘Ur’ refers to the origin or source of a case. Initially, the terms ‘oorzaak’ and ‘Ursache’ were used in devote texts, where, for example, God was described as the cause, origin or source of everything.

[2] A causal process is always a process related to the causal origin. In the effect system, there can also be a process that results in the effect being triggered by the causal process.

[3] The term ‘agent’ might suggest that an agent is a human or a human-like being, but that is not intended here. In the same way, acting is not only something humans do, but this can also apply to non-living, physical things.

[4] In a later publication, Byerly (1990) prefers interaction causes over substantial causes; see also the earlier quote of Byerly (1990). Byerly (1979) also uses the term effect system.

[5] In addition to the term ‘substance’, Byerly (1979) also uses the term ‘thing.’

[6] Baron-Schmitt calls α and β act types and are what Hornsby (1980) calls things done

[7] According to Dennett (1991; p. 402) shadows are causally effective: “your shadow has its effects on photographic film, not to mention the slight cooling of the surfaces it spreads itself over.” However, the mentioned effects are due to the light rays (emitted or reflected by a light source) having different access to surfaces.

[8] What is produced in the creative process of transforming the ingredients in an art object can be an Aristotelian substance, but not necessarily so. Arranging, e.g., a collection of stones in a particular pattern, does not result in an Aristotelian substance, although the arrangement of the stones can be an art object; see, for example, The Planet Circle and Bolivian Coal Line by the British Land Art artist Richard Long (https://depont.nl/en/collection/artist/richard-long).

[9] As we will see, I apply the notion of normality differently than McGrath.

[10] In Leuridan & Lodewijckx (2020), it is explained that in the pendulum case, two events are involved by assuming Kim’s property exemplification account of events. My reservations about this account were expressed in my blog post, ‘Is the death of Socrates the cause of Xanthippes becoming a widow?’. One can argue that my critique here is unfair because I changed from the property period to the dispositional property, of which the period is a manifestation. But then I would react that period is not even a property of the pendulum but of the movement of the pendulum.

[11] In contrast to what I am claiming, Place thought that “the macro/microstructure of an object stands as a cause to its dispositional properties as effect” (Place, 1996e, p.115). His reasons are hard to follow; I will say more about this in another context.

References

Anscombe (1971). Causality and determination. Cambridge University Press. (Reprinted in Sosa, 1975)

Armstrong, D. M., Martin, C. B., Place, U. T., & Crane, T. (Ed.) (1996). Dispositions: A Debate. Routledge.

Aronson, J. (1969). Explanations without laws. Journal of Philosophy, 66, 541–557.

Aronson (1971a). On the grammar of ‘cause’. Synthese, 22, 414-430.

Aronson (1971b). The legacy of Hume’s analysis of causation. Studies in History and Philosophy of Science, 2, 135-156.

Baron‐Schmitt, N. (2024). Thing causation. Noûs. doi:10.1111/nous.12494.

Byerly (1979). Substantial causes and nomic determination. Philosophy of Science, 46, 57-81.

Byerly, H. (1990). Causes and Laws: The Asymmetry Puzzle. In PSA Proceedings of the Biennial Meeting of the Philosophy of Science Association (Volume 1, pp. 545-555). doi:10.1086/psaprocbienmeetp.1990.1.192731

Cartwright, N. (2017).  Can structural equations explain how mechanisms explain? In H, Beebee, C. Hitchcock, & H. Price (Eds), Making a Difference: Essays on the Philosophy of Causation. Oxford Academic. doi:10.1093/oso/9780198746911.003.0008.

Dennett, D. C. (1991). Consciousness explained. Little, Brown & Company.

Fair, D. (1979). Causation and the flow of energy. Erkenntnis, 14, 219-250.

Hempel, C. G. (1965). Aspects of scientific explanation and other essays in the philosophy of science. The Free Press.

Hornsby, J. (1980). Actions. Routledge & Kegan Paul.

Leuridan, B. (2012). Three problems for the mutual manipulability account of constitutive relevance in mechanisms. British Journal for the Philosophy Science, 63, 399–427.

Leuridan, B., & Lodewyckx, T. (2021). Diachronic causal constitutive relations. Synthese, 198, 9035–9065. https://doi.org/10.1007/s11229-020-02616-0.

Mackie, J. L. (1979). Mind, brain, and causation. Midwest Studies in Philosophy, 4(1), 19-29

McGrath, S. (2005). Causation by Omission. Philosophical Studies, 123, 125–148. doi:10.1007/s11098-004-5216-z

Pearl, J. (2009). Causality: Models, reasoning, and inference (Second Edition) [First Edition is from 2000]. Cambridge University Press.

Place, U. T. (1973-1974). The metaphysical foundations of empirical psychology [A series of twenty-eight lectures presented to the Vakgroep Methodenleer, Sub-Fakulteit Psychologie, University of Amsterdam between September 1973 and May 1974]. https://utplace.uk/Amsterdam

Place, U. T. (1996e). Structural properties: categorical, dispositional or both? In D. M. Armstrong, C. B. Martin, U. T. Place, & T. Crane (Ed.) Dispositions: A debate (Chapter 7, pp. 105-125). Routledge.

Sosa, E. (Ed.) (1975). Causation and conditionals. Oxford University Press.

Woodward, J. (2003). Making things happen: A theory of causal explanation. Oxford University Press.

Leave a Reply

Your email address will not be published. Required fields are marked *