While thinking about the Butterflyeffect I came across some Questions.

First I want to mention that I want to exclude the general layer concerning determinism and indeterminism.

I therefore want to start my question by assuming that "Reality" is not fully predictable for humans due to incompletness of knowledge in humans and the complexity of "Reality".

I further assume we have a passing of time leading to different states and substates the world is in.

Furthermore I want to limit my question(s) to the temporal cause and effect relation.

I think classically the cause effect relation can be understood/conceptualised in a deterministic fashion. I am thinking along the lines of that a cause effect relation gets postulated when we observe that Event E1(cause) causes(go together with) E2(effect) reliably over many instances. Deducing from that we get that E1 necessarily leads to E2.

However given the first assumption we are in a probabilistic space. To illustrate the problem arising consider a coinflip. If we see E2 as one side the described necessity of E1 causing the specifc side is not given. If we see E2 as sum of all possible results we have the problem that due to the first assumption we can never consider all possibilities. F.e. In the coin example humans typically only consider the options that there will be a side on top and not that the coin could also stand on the side. We could maybe state that if a single outcome has a certain percentage or is one of the most likley events we define it as effect and therefore define effect/noeffect with a fuzzy border.

If we now focus on the Butterflyeffect we see that the flapping the wings of a butterfly leading to a tornado has a very low likleyhood. So it wouldn't be a effect in both suggested understandings of the terms. Especially in the circumstances of the Butterflyeffect it seems to be the case that not the event we view as cause is the origin of the phenomena and rather the structure of the enviroment/system the event takes place in.

Questions: 1) Does anyone know literature/authors regarding this topic? 2) Did I make specific mistakes? 3) Can/Is a classic view of cause - effect being defended. What are rebuttals? 4) Does the probabilistic approach create a contradiction with the terms as they are currently used. (Are the terms deterministic in nature) 5) Is it necessary in order to describe the relation between two events to add an additional component(f.e. systemstructure). (Do cause - effect need a logical connector).

Maybe additionaly: Is science that tries to explain the world via simple concepts(basic laws of nature/smallest particles) able to integrate concepts like the "sytemstructure" that are rather holistic?

I hope my Questions are not asked to open and I would already be happy if someone could answer 1). Thank you in advance.

  • What you describe is not classical causality. An event does not cause anything by itself, just think about the long chain of other events and surrounding circumstances that have to occur after killing the butterfly. Our singling out some event as "the cause" is just for rhetorical emphasis. You discovered that, taken literally, this naive rhetoric is not very coherent. Classical causality is formulated in terms of deterministic laws, and the butterfly effect is just sensitivity to initial conditions. – Conifold Jan 6 at 10:22
  • +1 However, there may be too many questions here. You could break these down into many separate questions asked over a period of many days. Regardless, welcome to Philosophy. – Frank Hubeny Jan 6 at 12:49
  • @MauroALLEGRANZA I already read Hume on cause and effect. I should have mentioned that. – CaZaNOx Jan 6 at 14:17
  • @Conifold I admit I didn't put it precisley enough. What I refered to with classical is it being deterministic and a effect being a necessity after it's cause has appeared in the chain of events. I just personaly think that you can reasonably reduce the matter of discussion by only looking at two elememts in the chain. If you want you can view both E1 and E2 as pattern combining multiple points in time but having a temporal order such that forall E1<E2. I am aware that determinism implies such a structure. Thats why I am asking about probabilistic models. – CaZaNOx Jan 6 at 14:34

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.