“From Plato to the present, with a few noticeable exceptions, reason rather than emotion has been regarded as the indispensable faculty for acquiring knowledge… Objective testability became the accepted hallmark of natural science; this in turn, was viewed as the paradigm of genuine knowledge… Because values and emotions have been identified as variable and idiosyncratic, the objective standard stipulated that trustworthy knowledge could be established only by methods that neutralized the values and emotions of individual thinkers.” (Alison Jaggar)

 

Jaggar is correct in stating that natural science has become the “paradigm of genuine knowledge” through its “objective testability.” I would argue that our society is intellectually and culturally founded upon scientific discovery. However, there are many methods of deriving knowledge which, while appealing to reason instead of emotion, are entirely separate from scientific enterprise. And, while society in general may hold the scientific method to be the paradigm, there are many commonly overlooked difficulties with the scientific method which render it much more subject to values and emotions than Jaggar’s understanding allows.

In discussing some of these roadblocks to scientific knowledge it is necessary to roughly outline the process of procuring knowledge scientifically in order to see how these issues fit into the broader scheme of the methodology of science. Data is the basis of science, which is what allows for the initial objectivity of the process. This data is used to form a hypothesis or a general theory surrounding the state of affairs behind the data. From here, scientists seek to demonstrate that observed data is consistent with and can, at times, be predicted with the theory. Once enough scientists have tested the theory, amassing a larger and more diverse set of data, the scientific community tends to regard the theory as true (at the very least operatively). 

Jaggar points to objective testability as the basis for natural science being considered the proper source of knowledge. However, the idea of testability needs to be qualified for any consistent definition of knowledge. Granted, these tests are objective, as they often deal with hard data. But how many times must something be tested and found accurate before one is able to reasonably assert a theory as truth? This question demonstrates the inductive nature of scientific progress. Data is obtained, and from the data, scientists reason abductively to create a hypothesis. Then, this hypothesis is tested, and the truth or consistency of the hypothesis is determined inductively. Yet, inductive truth is never absolute, rendering the “paradigm of genuine knowledge” probabilistic at best as well as entirely arbitrary with regard to how much testing must first be done. 

Further, as Hume noted, there are fundamental issues with inductive reasoning. Primarily, it begs the question. To reason from the past to the future is an example of inductive reasoning, the validity of which one can only hope to justify experientially. Yet, any experiential defense will first assume the abiding truth of one’s experiences, which is an application of the exact form of reasoning they are seeking to justify. This presents two issues; one in theory, and one in application. If one concedes that theories which are or have been true will remain true, there is still an incredible, often near infinite, amount of data that must be considered before a theory can be verified deductively. Because there is no hope of attaining or processing all of this data for the majority of important overarching theories, the consistency of a theory with relevant data must be reasoned to inductively from a limited data set. Within the realm of natural science, especially within a naturalistic worldview, what hope is there of justifying this inductive appeal? So, this form of induction, which utilizes a constant or momentary set of data, is stifled by Hume’s problem of induction. 

However, even were one to justify this aspect of the process, it still relies on the assumption of the constancy of data and/or theories, which is another manifestation of inductive reasoning. If, on a timeline, the aforementioned issue deals with “vertical” induction, this issue is one of horizontal or temporal induction. It deals with the abiding validity of theories rather than their initial, momentary, retrospective verification. Once one by some means verifies a theory, what justification is there for any belief in its continued truth? Perhaps one might suggest continued testing, which brings back the first issue. But this entails the idea that any moment in which a scientist is not testing or aware of others testing a previously verified theory is one in which the scientist has no justification for belief in it. On the other hand, what our scientific culture has chosen to do is to, once again, reason inductively and assume the abiding truth of these theories. This is yet another appeal to something that cannot be justified.

The problem of induction pervades the scientific process. This is because the scientific process is fundamentally inductive. But this inductive reasoning brings many questions surrounding epistemic warrant. How much data is needed? How many times must a theory be tested? How diverse must the data be? How often should a theory be re-verified? These questions, in society, reduce to psychological questions of confidence. It seems that this aspect of science, then, does bring values into the discussion. Jaggar differentiates between objective testability and variability, but these questions are riddled with variability. This is because the nature of inductive reasoning is not absolute. It can’t be, and Hume presented a strong argument for why it is not even valid. 

Because of this, many have tried to present scientific enterprise as a form of other types of reasoning. Some will readily claim it’s deductive, but any presentation of the method in a positively deductive form will be fallacious. Specifically, it affirms the consequent. If Theory A, Prediction B. Prediction B, therefore Theory A. This is, deductively speaking, entirely invalid. Yet, it is the closest to a syllogistic form of the inductive reasoning science employs that one can get. Others, recognizing this, present it as a form of negative deduction. Karl Popper, for example, holds a fallibilist understanding wherein scientists form theories and then seek to prove them wrong. Here, the form becomes: if Theory A, Prediction B. Not Prediction B, therefore not Theory A. This is logically valid. But is this really how we operate? Under this view, there is no way to prove theories, and all theories are equally justified so long as they are not yet disproved. Is it good enough that I haven’t yet died by eating a certain food, or should I have some reason to believe I won’t die? 

Yet, the same issue occurs with inductive views of science. All theories which are consistent with the limited data set that is available must be held to be of equal viability. Empirically equivalent theories cannot be compared through these inductive means, and thus neither can be held as knowledge. Yet, if there is only one posited theory that is consistent with the data, many will hold it to be knowledge in spite of the potential for other theories. This is the inherent weakness of scientific enterprise as the basis of knowledge. Because it is inductive, it requires the values of scientists to be introduced through their interpretation of the data. 

That is not to say that science is useless; it is very helpful given the uniformity of nature. However, it is not the only source of knowledge which elevates reason above values and emotions. In fact, it is not even close to pure deductive reasoning or even non-scientific reasoned abduction. It is silly for scientific enterprise to be held as the paradigm of genuine knowledge when there is not a shred of absolute knowledge to be had from the scientific method. It is left up to the individual to decide his own epistemic norms. These are not moral or religious values, but they are still values. 

So is there any redemption for science? Certainly; science has a holy and proper place in our ventures to understand God’s creation. Hume’s problem of induction is not an issue within a Christian worldview; we don’t make appeals to experience. We don’t need to. Rather, God’s revelation affords us a transcendental appeal. Given Christianity, God’s infallible word gives us sufficient reason to believe in the uniformity of nature.

“Thus says the Lord: If I have not established my covenant with day and night and the fixed order of heaven and earth” (Jeremiah 33:25)

Because of this revelation, Christians can pursue science boldly and faithfully. Science is not to be considered the domain of the secular; rather, it is Christian capitol that unbelievers have borrowed when they engage science. There is no single body of “science.” There are only scientists. Some are faithful believers in Christ, and others are idolaters. There is no unbiased, objective scientific truth, because the scientific process is not objective by any measure beyond the collection of data. The moment human interaction with and understand of the data enters the scene, sin follows, as well as a number of other cultural, political, and economic motivations.

Science is not opposed to Christianity. It is not wrong, and it has many strengths and benefits. However, it should never be placed on a pedestal, and it must always be subject to the revelation from God which first justifies it.

Grace and Peace