1) All animals seem to function on the assumption that the future will be like the past. Otherwise, the uncertainty which would ensue would hinder the survival of even the most capable of animals. The rational animal has developed a magnificent theory which accounts for this phenomenon; namely, the theory of causation.
2) Causation is the theory that all events in the world are the result of some cause. Insofar as the same causes are in place, the effects will necessarily be exactly the same. Determining which causes are responsible for which effects is the occupation of science. This is achieved through experimentation, so as to isolate particular causes and their effects.
3) Since every event must have a prior cause, which itself must be caused, and so on ad infinitum, the chain of causes must have commenced with a first cause, which itself is uncaused. This idea lies at the heart of religious philosophy, in which God is ambiguously taken to be the uncaused first cause.
4) The apparent success of this theory has allowed it to remain virtually unchallenged for centuries. Eventually, the keen eye of philosophy turned its attention to this fundamental, and it was immediately revealed to be groundless and wholly speculative. I am referring, of course, to the works of David Hume in the eighteenth century, particularly An Inquiry Concerning Human Understanding. Hume simply asked, what informs us of this connection between two events? Can the two events not be distinct and unrelated, each occurring uncaused, as a matter of fact? This is known as the problem of induction.
5) Intuitively this seems wrong. The fact that events have always conjoined in this way (at least so far as we know) indicates this phenomenon. Why else would two events always succeed each other so precisely?
6) However, this begs the question and does not remove it. For if there is no inherent connection between two events, there is no ‘why’ question which is valid. If it is merely a matter of fact, on what basis can one ask ‘why’ the fact is this way, and no other? Only upon assuming a worldview of causation can questions of the ‘why’ sort have validity.
7) This is the failure of the axiomatic question (commonly posed by theists) as to why there exists something rather than nothing. Such a question assumes that there is a ‘why’ for the existence of reality, but this is a baseless assumption. In fact, even if causation is a viable thesis, this only tells us that in the world events function this way, but the existence of a world which functions this way in the first place cannot be restricted by the contingent mechanism of the world, since these limitations only exist if the world exists. Moreover, assuming this world encompasses all of reality, the existence of this world cannot be dependent on anything, including causation, for without the world there is nothing which can contain and dictate such limitations.
6) No occurrence in nature can disprove causation, but can only disprove our understanding of particular causes. Indeed, similar events in nature are constantly succeeded by slightly or entirely different events, but this does not represent a problem for the believer in causation. If a different succession has been obtained it is determined that there is some difference between the two prior events, which is the reason for the differentiation.
7) Given certain occurrences, the laws of nature, which are generalizations of induced causes, may have to be rewritten; but the assumption that there are laws still holds, because anything which occurs is necessarily a part of nature, and so must be included in, and accounted by, the generalizations of nature. This is often overlooked by religious laymen who see in supposed miracles a proof for the existence and authority of God. What, though, is a miracle, if not an event in nature? Even assuming any such event has occurred, it can reveal nothing more than the necessity to revise our current generalizations of nature.
8) So it must be granted that empirical observation cannot confirm or deny that events are caused. Immanuel Kant explored an entirely different path to account for this precious belief of ours. He postulated that causation is not known to us through experience, but is known a priori as a given. Experience determines how we are to assign the presumed causes, but does not, and cannot, create the need for such assignment.
9) Kant was thereby attempting to restore security in this conviction, but this theory mostly serves to increase doubt in causation. After all, a priori reasoning isn't something demonstrable, and can therefore be wrong. Later academics, following the prevalent trend of accounting for everything in terms of its survival-merit, have postulated that this belief arose because it allowed man (and other animals) to accurately predict future occurrences. Such an explanation may be helpful to scientists, but it leaves logicians embarrassed and unemployed.
10) Other explanations were theorized as well. Hume himself accounted for this belief rather radically. He argued that belief and imagination are of the same kind, different only in degree; belief arises when there is high expectation for a particular imaginary state. Given this supposition, he postulated that we believe the future will be like the past out of habit. Our brains, accustomed to seeing event B succeed event A, expect by force of habit that it will continuously be this way.
11) However, this explanation only accounts for the belief that the future will be like the past, but does not account for the belief in causation. Personally, I am inclined to believe that it arises due to the nature of our rational faculties. To say events are uncaused, that they are merely a matter of fact, does not explain the fact but insists that no such explanation exists. This does not satisfy our rational faculties, which desire an explanation, and so causation is theorized. Accordingly, it is not from the observed frequency of similar successions which forms our conviction in causation, for there is nothing in the repetition of similar events that is not in any one of those events, as Hume demonstrated; rather, any single event is deemed caused in effort to make sense of it.
12) Such a phenomenon is apparent from the argument theists are accustomed to hold concerning the existence of God, that postulating a God explains the existence of the world better than not postulating a God. Yet, the atheist position is not that the world can be explained without a God (that would be absurd), but that no such explanation is necessary! The onus is on the theist to demonstrate the need for an explanation. It seems obvious that such argumentation is the unfortunate result of our need for explanation, owing to our rational faculties, which desire explanation. Like a child who must know ‘why’ to everything told to him, the theist hangs onto God to satisfy his need for explanation.
13) Truthfully, any explanation here is unnecessary and contradictory. Unnecessary, because if events are not caused, the event of our believing in causation or the conformity of the future to the past can also be uncaused, and so does not require an explanation. Contradictory, because it purports to establish causes in order to maintain randomness. Apparently, Hume did not recognize the implications of his argument even while introducing it.
14) We do not believe in causation and therefore deduce that the future will be like the past, if anything, it is the belief that the future will be like the past which necessitates the belief in causation. This can easily be demonstrated. We believe that the world will continue to exist as it did in the past, despite atheists and theists alike not positing any necessary cause for the existence of reality. It is thus evident, that this belief is independent of the belief in causation; it is merely made coherent through it. Consequently, if we are to end this period of philosophical frustration which Hume has begun, we should not look in causation for our rationale, since it is not causation which is the source of this belief.
15) How is it that any belief can ever be justified? Even assuming heaps of evidence and sound argument, it is possible that there is something we are unaware of which accounts for the evidence and arguments we may have. After all, we cannot disprove that which we are unaware of, and we cannot prove that we are ever sufficiently aware to make claims.
16) We mustn't consider such objections to our beliefs, because beliefs are established through judgment, which is a subjective act based on our perspective. Beliefs are justified so long as they concur with the knowledge available to us, because beliefs are this subjective judgment. Nonetheless, this is a strong objection to knowledge, as it makes us aware of our indispensable fallibility.
17) When we observe the succession of events A and B, and then inquire what would happen if we were to repeat event A, we believe event B will follow because this is the only knowledge available to us. To entertain the notion that event A might be followed by event C is to entertain the notion that something new would exist, namely event C, which as of yet we have no knowledge of. Even if event C consists of nothing but the lack of event B, that as well is a state of affairs which we have no knowledge of its existence. Therefore, we judge based on the knowledge available that event B will follow, as opposed to event C. This is the source for the belief that the future will be like the past, and is its rationale; the future is determined to be like the past because to say it will be different than the past is to posit the existence of that which we have no right to posit. With this simple idea I believe the problem of induction is solved.
18) No doubt one will object to this, that the repetition of event A is not the same as the previous event A, so what will follow thereafter cannot be learnt from the previous event A. Any repetition of event A is actually a new event D, so no knowledge of what followed A can be of any assistance. At first glance, this is a strong objection.
19) However, to say that a repetition of event A is not the same as the previous event A, but is rather a new event D, is itself an invalid assumption: because until the existence of event D is known, we cannot rationally assume that this is a new event D, rather we must assume that this is but event A again. The fact that it consists of different matter and occupies a separate time and space is not reason to conclude that this is a new event D, after all, it can still be event A, but a different manifestation of it! This is not to say that by witnessing the repetition of event A we do not learn of anything new, we certainly do. Nevertheless, nothing which we can learn in this new occurrence can show us that this is not event A, but an entirely new event D.
20) The probability of an event X as opposed to an event Y is found in the results obtained in similar conditions. In any particular event probability does not exist, it either is X or Y. This is obvious if we define probability as measuring the actual results of similar events, as many philosophers do. However, it is also true if we define probability as measuring the propensity of the conditions to produce those results, as some philosophers do. Even propensity-theorists refer to different events to find probability; they only add that the actual occurrence of those events is irrelevant, probability according to them is based on the physical conditions responsible for those results.
21) If probability only informs us of different events, as we just established, how can we use it to inform us of the result of a particular event? A particular event in the series can either be X or Y, how does the knowledge of other events inform us anything about a particular event? Yet, we constantly use probability to determine the likelihood of an event, this assumption is fundamental for much of our knowledge.
22) The concept of probability is generally understood to be intimately connected with the concept of causation. If events were random or uncaused probability wouldn't be measuring anything, and past results wouldn't tell us anything about the future. Unfortunately, this conception of probability makes it even more difficult to understand our rationale in using probability. Take for example the probability of a coin-toss landing heads or tails. Now, if everything is determined that means every time a coin lands heads or tails it is because of some particular cause applicable only to that case- so how can we measure probability if every case is unique? The outcome of any coin-toss depends on particular predetermined factors unique to that coin-toss; on what basis can we measure the probability of coin-tosses in general?
21) According to the above explanation of induction, we can account for probability as well, without appealing to causation. Probability arises when we have observed different successions of an event A, sometimes B and sometimes C. Thus, when faced with the repetition of event A we use the knowledge available, that X amount of times it was succeeded by event B, and Y amount of times it was succeeded by event C to determine that for this repetition of event A there is X probability that B will occur, and Y probability that C will occur. To propose any differently is to propose the existence of that which we don’t know exists, when we can instead account for this succession with what we already know exists, which is an unjustified proposal. For this reason we use probability despite it only informing us of other events, for these events are assumed to be related so long as their distinctness isn't established.
22) Occam’s razor, the principle in logic which states that when faced with competing hypotheses one should prefer the simplest one, meaning the one involving least assumptions, is based on the above expressed idea of unwarranted assumptions. If a simple explanation accounts for phenomena, then a more complex explanation, which grants more assumptions, is unwarranted and groundless, albeit theoretically possible. In order for a hypothesis to be justified, it must follow from the knowledge available to us alone.
23) With this we can turn to the justification of positing causation, for as of yet we have only justified the belief in the conformity of the future to the past. Causation is the theory of events which grants the least assumptions, and is therefore assumed. To posit that events are uncaused is to make every event distinct and unrelated, each occurring as its own reality, whereas, to posit that events are caused is to make numerous events related, each occurring as an expression of one reality, namely, the cause. Thus, this assumption is justified because of its simplicity.
