Is it Falsifiable?

What is science? This is a question that scientists, philosophers, and others have discussed with no definitive conclusion. Common definitions of science require that a scientific idea be testable, falsifiable, and predictive. Many of the definitions of science are developed in order to distinguish science from pseudoscience. Pseudoscientific ideas are those that may appear to be scientific but fail some of the basic requirements of a scientific theory. They may be unfalsifiable, their proponents may consider only data that supports the theory and none that contradicts it, others may not be able to reproduce any of the results that might confirm the idea, or other such problems.

Although the claims of Christianity are not scientific claims, per se, the same tests that are used to help confirm the truth of a scientific theory can be applied to religious claims. Over the last several blog posts I have shown that the claims of Christianity can be scrutinized using some of these same criteria used to test a scientific theory and can be shown to have validity. For instance, the claims of Christianity do have external confirmation, can deal with counter-arguments, and are logically self consistent. In this blog post we address the seventh of eight criteria used to assess the truth of any particular hypothesis: "Can the hypothesis be falsified or confirmed with other data?"

External Confirmation Required

At every laboratory where I've done research there have always been at least two major experiments designed to investigate similar scientific questions. At CERN, where I currently do research, I am a member of the ATLAS collaboration, one of the two "general purpose" experiments at the Large Hadron Collider (LHC) constructed to explore a broad range of scientific questions with LHC data. The CMS experiment, which sits at the opposite side of the 27 km (17 mile) circumference tunnel, is also a general purpose detector designed to look at a broad range of scientific topics. Despite the fact that having two somewhat redundant experiments costs twice as much to build and maintain, this arrangement is optimal so that each experiment can corroborate the results of the other. New discoveries and measurements require external confirmation to affirm their validity, and so complementary experiments are established in order to provide the necessary verification.

There have been a few times during my career in particle physics that one experiment seemed to have evidence for a discovery of something entirely new, but was eventually shown to be wrong, partially because other experiments were unable to provide external confirmation. Such cases involve the false "discoveries" that quarks have substructure, that particles can travel faster than the speed of light, and that weird particles called lepto-quarks actually exist. (These things may still turn out to be true but the past experiments that seemed to have found them have all been shown to be incorrect.)

External confirmation is not only one of the requirements for determining if a proposition is valid or not in any scientific endeavor, but also in other arenas where claims about objective truth are made. In a series of blog posts I have been applying some of the same principles used in my scientific research to the beliefs and world-view of Christianity to investigate whether or not they seem to have objective validation. I have already addressed the questions (1) "Is the data logically self-consistent?", (2) "Is there enough evidence to support the hypothesis?", (3) "Is the hypothesis compatible with other known data?", (4) "Is contradictory evidence conclusive?", and (5) "Is something essential missing?". This blog post will address the sixth question, (6) "Is there External Confirmation?", while two future posts will discuss the final two questions,  (7) "Can the hypothesis be falsified or confirmed with other data?", and (8) "Are there other possible explanations that are more feasible?".

Is Something Essential Missing?

In 1989 two chemists, Martin Fleischmann and Stanley Pons held a press conference to announce that they had discovered a process in which nuclear fuel would fuse together to create heat and energy in a small tabletop experiment. Their supposed discovery of "cold fusion" held out the hope of a cheap and abundant supply of energy for the whole world. Yet there were some immediate problems with the experiment. For instance, other scientists were not able to replicate the results despite following the recipe given by Pons and Fleischmann. But perhaps the most obvious hint that something was just not right was the absence of neutrons. In every type of nuclear fusion or fission, excess neutrons should be released. Pons and Fleischmann had originally claimed that their experiment produced excess neutrons, but when other experimenters saw none it became clear that they had not produced any either. When nuclear reactions occur, neutrons must be present. If they are missing then nuclear reactions are not taking place.

Over four previous blog posts, I have been discussing questions that scientists sometimes ask to determine whether or not a particular claim is true, and then asking those questions about the truth claims of Christianity to evaluate its veracity. One of those important questions is, "Is something essential missing?" If a certain required element is missing from a proposed explanation, then the explanation is likely to be not true. When it comes to truth claims from many of the world's religions and various world-views, I believe that there is an essential element that is present within Christianity, but missing from many other philosophies; thus giving credence to the Christian world-view. The issue has to do with human nature and human actions.

Challenge to Christianity: The Problem of Evil

Usually when scientists do multiple experiments to test a theory or model, there will be some experiments that seem to be just a little bit outside of the model predictions. There are almost always a few experiments that are in slight conflict or contradiction to the expectations, even when the model is otherwise well established as being accurate. In science these outlying experiments are often simply due to the nature of statistical analysis. For instance, one out of ten experiments is expected to be about three standard deviations from the average. Sometimes an outlier occurs because the measurement is not very accurate, so it appears to be far from the average but is still sensible given the uncertainty. Sometimes we don't have enough data to make definitive conclusions. In any case, it is important to evaluate if contradictory measurements have reasonable explanations or if they are a real problem for the model being tested.

This is the fourth blog post in which I am applying some of the same criteria that scientists use to test whether or not a theory is true to the claims of biblical Christianity in order to gain insight into whether or not Christianity has objective credibility. The first three questions that have been addressed were, (1) Is the data logically self-consistent?, (2) Is there enough evidence to support the hypothesis?, and (3) Is the hypothesis compatible with other known data? In this post we ask the question (4) Is contradictory evidence conclusive? Nothing is known to 100% accuracy, and every theory or model has some evidence that may appear to contradict the theory, those outlying experiments. But an idea that is valid will have reasonable explanations for those deviations so that the model remains credible.

One of the major challenges to Christianity is the problem of evil and suffering: how could a good and omnipotent God allow evil and suffering? I believe this may be the hardest philosophical question to answer, not just from a Christian worldview, but from any worldview.

Explanatory Power of a Biblical Worldview

As a professor of physics at a research university, I will often get unsolicited papers from non-scientists that propose an entirely different perspective on some known physical phenomena, maybe the electromagnetic force, or how atoms behave. In almost every case I can quickly glance at the paper and recognize that the proposal cannot be true. I may be accused of being arrogant or biased because I quickly dismiss these new ideas with apparently very little thought. But, in reality, most of these ideas can be readily discarded simply because they contradict known data and experiments.

For example, a student once came into my office with an idea that he had been working on since he was very young about how electricity might be described and modeled. It was apparent that, although his idea was clever and inventive, it did not coincide with some of the things we know about the electromagnetic force. Rather than squelch this student's idea outright I simply informed him that since he was a freshman physics major he would learn a lot about electricity during the next four years of college. I suggested that he compare what we know about how the electromagnetic force works with his model and see if his model was able to accurately describe what we know from experimental observation. If it didn't, he'd have to revise or reject his model. If it did, his model may be correct.

In any effort to determine if an idea is true or not, it is vital to compare the predictions of the idea with what is already known to be true. If an idea contradicts known truth, then the idea cannot itself be true.

Evidence for Christianity

If you flip a coin a number of times and 60% of the time the coin shows "heads" while 40% of the time the coin flip shows "tails" would you conclude that the coin is a fair coin that is properly and unbiasedly weighted? As the question is stated, there is not enough information to accurately answer it. For instance, if you flip the coin ten times, it is quite probable that the coin will be heads six times and tails four times. A six to four split should happen about 21% of the time, so such a pattern would not be unusual in 10 coin tosses. But suppose instead you flipped a coin 1000 times and it showed heads 600 times and tails 400 times. That is still a 6/4 ratio but the probability of this pattern appearing is 4.6×10^-11. Clearly a coin showing 600 heads and 400 tails is not a properly balanced coin. In order to make a definitive statement about the coin there must be enough data, or evidence, to support the hypothesis. (In reality, it seems to be impossible to make a "weighted" coin by adding weight to one side, but a coin can be made to land on a preferred side by bending it.)

This is the second installment in a series of blog posts dealing with some general questions that can be used to try to determine if a certain proposition is true or not. These principles can be applied to a scientific experiment or to the claims of Christianity to ascertain what conclusion is likely true. In a previous blog post I asked the question of whether or not the data was logically self-consistent. Regarding Christianity, I claimed that the Bible itself is a self-consistent book, and that the message of Christianity regarding the person of Jesus, the character of God, and the nature of humans is also self-consistent.