This basis of this post has been brewing for quite a while, but it didn’t seem like it would make a long enough post to actually do until now. This will have two parts, a brief discussion of the two common types of arguments and how they are misconstrued, followed by a discussion of which one science uses and why that is significant. So, let’s get started.
Most people have been taught that the difference between deductive and inductive arguments is that the former is a type of argumentation that moves from general principles to specific application, while the latter moves from specific instances to general principles. This is probably not news to anyone, I even found this categorization in an otherwise really good Christian educational psychology book Created to Learn.
I say otherwise good, because unfortunately, this is not actually the difference between the two types of argumentation. It is possible to create deductive arguments that move from specific to general and likewise is it possible to create inductive arguments that move from general to specific. They do in fact generally follow the above schematic, however, since there are exceptions, there is most likely a better explanation for the difference between them.
The difference, more accurately, has to do with the probability of the conclusion. In a deductive argument, the conclusion necessarily follows. In an inductive argument the conclusion only probably follows. So in other words, a deductive argument will always be a stronger argument, for if its premises are true and reasoning valid, it produces a sound argument that is rather airtight given the truthfulness of its premises.
The misfortune that most people are not aware of this, probably (reasoning inductively here) stems from the fact that most people do not formally study logic as part of their education. This is especially a vital study for people planning on entering ministry, yet very few pastors actually engage in a formal study of logic. Even my pastor committed a fallacy in reasoning this past Sunday in a sermon (although it was somewhat forgivable and related to etymology of a Greek word, which did not affect his overall point, if you’re curious ask me about it).
This is also why we have things like inductive bible study, which is based on the fallacious understanding of inductive arguments. While you may think that in an inductive Bible study you are gleaning principles to then stringently pattern your life around, those principles, if you were in fact reasoning inductively, only probably followed from the observations you made from the text, they did not necessarily follow like some people pretend that they do.
Not necessarily a reason to doubt what you glean from an inductive Bible study, but definitely a word of warning to not be too certain in your conclusions and always be open to re-evaluation as you grow, and also to not become legalistic in following your principles that may not be as transparent in the text as you feel that they are.
Anyway, all of this was to establish that difference between reasoning inductively and reasoning deductively has to do with how necessary the conclusions are. The reason this is important for where we are going in the larger context of this blog, is that science does not have a deductive method of proceeding. Science can at best offer inductive arguments for its ideas, but it cannot offer deductive ones, or at least it doesn’t in practice.
Science rather, proceeds usually by either one of two ways. Either it uses Emunerative Induction (henceforth E.I.) or it uses the Hypothetico-Deductive Method (henceforth HDM) which despite the name is not deductive, and actually isn’t even inductive either.
E.I. is something we all use, and in fact the basis of your sense experience rests on your use of E.I. That and your unprovable assumption in the uniformity of nature (natural laws are the same across all time and space). Basically, E.I. operates by repeated observation that under stable circumstances, an event repeatedly occurs. For example, you find that every time you kick a rock while barefoot, a painful sensation occurs in your big toe. Without fail, every time in the past this happened, it hurt like hell (if you’ll pardon the expression). You induce then, that any time in the future, this is probably what will continue to happen. (Note the use of “probably”)
This is how the aforementioned natural laws are formulated and how we understand the uniformity of nature, which should immediately alert you that neither are predictive in nature but are merely observations of what has been true in the past. In other words, there is no guarantee a law of nature will not fail at some point (or could be violated at some point), it is just unlikely given that it has not failed (to your limited knowledge) in the past.The uniformity of your past experience does not establish that uniformity of the present (which is itself just a different use of E.I.).
The inductive nature should be clear, in that nothing is absolutely certain. From the repeatedly experience of stubbing your toe, it does not necessarily follow that in the future this always be true, just like from the repeated evidence that gravity has worked in the past, it does not necessarily follow that it will always be the same in the future. In both cases it only probably follows. And although it is a high probability, it is nonetheless an act of faith on your part (and the scientist’s part) that what you observe in the past will be the same in the future.
And there is the rub for the whole science vs. faith issue. One might think that HDM can save the scientist from ultimately having to operate on the basis of faith (albeit faith in a different object than the Christian, but still an act of faith). Unfortunately, HDM is basically a form of abductive reasoning, which technically speaking, is an invalid form of reasoning. The conclusions it renders may nonetheless be true, but it is an invalid way to get there.
To somewhat see why this is, it is important to realize that in science, data always under-determines theories. Another way of saying this is that no finite data set can conclusively prove a given theory. There is always another explanation for the given data, and in fact, there is literally an infinite number of explanations for a given data set. Most of them might be ruled out as improbable, however they cannot conclusively be proven false in any sort of deductive fashion. Nor can one’s cherished theory be proven conclusively true by a given set of data.
What seems reasonable, or probable, is ultimately a function of the background assumptions a particular person brings to the data set. All HDM is really saying is that if this particular theory is true, we would expect to observe this particular data. However, the presence of the data does not prove the theory. It supports the theory, but another theory could just as easily account for the data and then one has to decide which is preferable, but at that point, you are no longer dealing with the data. You are dealing with your presuppositions (background assumptions), and that is really where the rub is when it comes to science.
It is simply wishful thinking to believe that scientist operate in some sort of rationalistic unbiased fashion as opposed to theologians who operate based on faith and the words of a sacred text. In reality, scientists operate on faith and the “words” of a sacred text, they’re faith is just in different place and their “text” is the book of nature rather than the Scriptures. But they are still operating with background assumptions, they are still engaged in interpreting a “text” and they are still ultimately dependent on faith based assumptions.
To see this in action, all we need to do is look at Richard Dawkins’ latest book. And we’ll do just that tomorrow.