Why I am Convinced God Exists 2: Assumptions and Evidence Standards

Michael Angelo's Creation of Adam (Source: Wikimedia.org)

Michael Angelo’s Creation of Adam (Source: Wikimedia.org)

In attempting to answer the question “Does God Exist?” I had to make some assumptions. In this post I discuss the rationality of my process, and lay out the handful of assumptions that underlie my broader analysis. These assumptions are, I believe, entirely reasonable: 1. I exist. 2. I can reason. 3. There are true answers to many questions related to God’s existence. Based on these, I discuss specific standards of evidence and proof used by various disciplines.

Previous Post: Introduction

In my previous discussions, I’ve used the term rational a couple times, and so it’s important that I tell you what I mean by that. Rationality is a term used in economics and political science to describe a process of thoughtfully pursuing goals using reason. That is, a person is considered rational when she has a goal she’d like to achieve, is able to reason through a set of options for pursuing that goal, and follow a path she believes will help her achieve that goal. This evaluation and path selection assumes she has enough information to make those judgments, and that she can reason through the consequences of selecting different paths. This doesn’t require perfect or complete information, but having enough credible information to make a reasonable prediction.[1]

Here, my goal is to conclude whether God exists, and rationality refers to my reasoning through the various paths that come up on that journey. As a result, I haven’t specifically addressed every question that could possibly be asked about what the world would be like if God doesn’t exist. Rather, I’ve selected the most pertinent lines of inquiry that I reasoned would help me answer the question for myself. Other people will probably have other lines of inquiry that will satisfy their intellectual curiosity on this question. If they’re rational, they’ll think systematically, too, and will pick questions that will, on balance, enable them to more or less objectively answer the question. The alternative is to cherry-pick paths that will lead them to the answer they want to be true, which is intellectually faulty and dishonest.

After I wrote about 12 single-spaced pages of analysis, I stopped. Had I made unreasonable assumptions that would inexorably lead me to the conclusion I wanted to be true? That is, since I want God to exist, did I frame the whole set of questions in such a way as to inevitably lead me to the conclusion that God exists? As I thought about it, I backtracked to my initial questions to figure out my underlying assumptions. I concluded that they are reasonable and realistic and need not lead to the conclusion that God exists. They are:

  1. I exist. By implication, this also means the material universe, what Carl Sagan called the cosmos, exists.
  2. I can reason. That means I have the intellectual ability to think through a series of questions and answers, imagine contrary views, evaluate contrary arguments, compile and evaluate evidence, and draw conclusions based on that process.[2]
  3. There are true answers to many questions. Some statements are true and other statements are false. By true, I mean true in the conventional way that people use the word and its partner, truth. The Oxford English Dictionary defines a true statement or belief as “Consistent with fact; agreeing with the reality; representing the thing as it is,” and truth as “Conformity with fact; agreement with reality; accuracy, correctness, verity (of statement or thought).”[3] So the true answer to whether God exists has to be consistent with the facts, agreeing with the reality of the way things are. Ascertaining the truth of a statement involves the exercise of reason (see my second assumption, above).[4]

That’s it—those are the core assumptions from which I’m working. Fundamentally, this is a search for the true answer to the question, Does God Exist?. Again, I don’t claim especially comprehensive or specialized knowledge in some areas, but I know how to evaluate many facts, I know how to compare and contrast ideas, and I know how to think reasonably logically.[5] I think I’m competent to evaluate whether an idea, concept, or statement conforms to reality within the normal bounds of normal human intelligence.

My process, therefore, was not to develop my own professional expertise, research agenda, or even a deep familiarity with the original professional academic research in any of these areas. Rather, I did my best to evaluate other people’s work while also thinking systematically through my own ideas and considering in broad strokes the alternatives. And that has to be enough. (Even if I wanted to acquire professional expertise in some of these fascinating areas, I already have a full-time job and my lovely bride has said she is not going to put me through graduate school again!)

Now that I have discussed my basic reasoning and assumptions, let’s turn to specific issues related to the standards of evidence or proof one might need in any area of study.

Standards of Evidence

Convincing someone of something depends on the standards of evidence the first someone requires. What kind of evidence would be adequate to prove that God exists? Logic, legal reasoning, experimental science, statistical reasoning and historical reasoning all provide input into addressing this question. In the end, I conclude that science and experimental reasoning provide language but not tools for evaluating whether God exists, because they are only appropriate for testing material things and processes. Since no one makes a serious claim that God is a material being, materialistic methods for “testing” for God simply aren’t appropriate.

There are many common standards used and familiar to many of us, and different disciplines make use of different proof standards. The “proof standard” leading to the conclusion that my ancestor Caleb Albee fought in the Revolutionary War is different than the proof standard that education affects the probability that a person will vote. The legal standard expected of juries for convicting criminals is different from the experimental standards for concluding that the rate of chemical reactions varies based on changes in the ambient temperature in which the chemicals interact.

Some common proof standards

Beyond a reasonable doubt. This is the burden of proof a prosecutor must meet in a trial in which someone is accused of a crime. The jury must find the evidence and reasoning so strong that any reasonable doubt about the accused’s guilt must be overcome. Note that this doesn’t eliminate any doubt, or any possible other explanation for a crime. Even though this sounds really difficult to understand, thousands upon thousands of our fellow citizens are able to apply this proof standard every year in nearly every jury trial held in our nation.

Preponderance of the evidence. This is the burden of proof that must be met for someone suing someone else in a civil suit, such as breaking a contract or agreement, or that the tree branch that fell on my car is your fault because the tree is on your property. Lower than the ‘beyond a reasonable doubt’ standard, this simply requires a jury to assess whether, when all the relevant evidence is considered, someone broke a contract or ought to be held responsible for some bad outcome. Again, thousands of our fellow citizens apply this proof standard in civil jury trials every year.

Reasoning to the best explanation. This is a proof standard for drawing conclusions about historical events and relationships, such as in my hobby, genealogy.[6] This leverages documentary and archaeological evidence about the past: events and people, their lives and relationships based on contemporary accounts, documents and circumstantial evidence. This is fundamentally the basis for the discipline of History, which seeks to document whether, where, how, and why certain events happened. Since for most of human history there are no eyewitnesses now living, everything we know about the past is based on reasoning to the best explanation. (I have recently come to learn that in Logic, this is called abductive reasoning.)

Scientific statistical reasoning. This is based on probability and sampling theories, in which the likelihood of outcomes and relationships can be calculated; and hypothesis testing.[7] There are normally two major types of errors to avoid: rejecting a true statement instead of accepting it as true (a Type 1 Error), and accepting a false statement as true instead of rejecting it (a Type 2 error). Probabilities can be calculated for avoiding those errors. (In terms of God’s existence, a Type 1 Error would be rejecting the statement “God exists” when it is in fact true. A Type 2 Error would be accepting the statement “God exists” when it is in fact false.)

  • The standard point at which a statement is normally accepted as empirically supported is at the 95% probability level, though that standard is arbitrary. You have seen this used when the news presents public opinion poll results that include a margin of error notation (such as ± 3%). Billions of dollars in spending each year hinges on people applying this proof standard to marketing and political decisions. The important thing about this kind of reasoning is that it hinges on a probabilistic standard of proof rather than an absolute standard; even if we obtain a result for which we are 99.9% sure, we still allow a small probability (0.1%) that our conclusion is wrong. The lower that probability, the higher the confidence that we’ve correctly identified a true statement or rejected a false one.
  • Explanatory power is the statistical concept that permits us to calculate the proportion of variation in some phenomenon that is explained by a set of independent variables. Any group of explanatory factors will explain some amount of the differences in a thing we’re trying to understand. In my research, for example, I often am trying to explain why people vote or do not vote. The outcome is easy to understand: either a person votes or she doesn’t vote. If I have 10 variables that are plausibly related to that behavior, I can calculate rather precisely how good a job each one does individually. I can also calculate how good a job they do collectively.[8] One important point here is that even in the most rigorous analysis of individual behavior we never even come close to explaining 100% of the variation in anything (aggregated social statistics are an exception).
  • So as a social scientist using statistical and econometric methods, I am comfortable with (1) finding explanatory factors that are plausibly, logically and statistically related to a phenomenon I’m trying to explain. (2) Thinking probabilistically about how likely they are to actually be related to the thing I’m trying to explain, and being comfortable with that probability being a good deal less than absolute certainty. And (3) happy when a reasonably high level of explanation is provided by my factors without even coming close to explaining everything. I am satisfied with sufficiency and coherence, rather than incontrovertible proof, and every social scientist and economist that uses statistics operates on the same principles.

Experimental evidence of physical objects and processes (that is, conclusions that are based on the process of hypothesis testing, controlled experimental manipulation, and replication). Since this particular proof standard requires the ability to manipulate and replicate experimental conditions, it is inadequate to evaluate claims that cannot be subjected to experimental replication or manipulation. For example, we can’t replicate the creation of something from nothing, in which there is by definition nothing to vary or manipulate in the experiment; thus creation ex nihilo is an event not subject to scientific experimentation. On the other hand, experimental evidence is adequate for learning all kinds of things about the way the material world operates.

Basic Logic, including the Law of Noncontradiction, which says that “No proposition may be simultaneously true and false” and its partner, the Law of the Excluded Middle, “Every proposition must be either true or false.”[9] (Note that is a more formal restatement of one of my underlying assumptions.)

Understanding these standards of evidence and proof helps me reason about God’s existence in three specific ways.

  1. First, I can be okay with not having 100% proof of God existing, but with finding sufficient arguments and evidence for me to conclude that some things are very likely to point to God’s existence. Admittedly, this is a bit subjective, and the question is not subject to actual probabilistic calculations at all! But the principle is that in the rest of my intellectual life I don’t expect or get absolutely perfect or complete explanations for anything, so I’m comfortable with some gap between what I successfully explain and a perfect explanation.
  2. Second, a variety of kinds of arguments and evidence can be brought to bear on the question of God’s existence. When I teach students about methodological variety, I use the metaphor of triangulation, in which methods and accumulation of different kinds of evidence provide a process of narrowing the set of potential explanations to those that seem most plausible.
  3. Third, given what I observe in the world around me, I can reason out which theory and factors best provide a comprehensive, if not complete explanation for the world and for human behavior. Do the principles of an atheistic worldview do as good a job explaining things as the principles of a theistic worldview? If not, which theism provides the best or most reasonable explanation, or demonstrates the greatest explanatory power, for the way the universe and human beings actually operate? Again, I don’t need absolutely complete explanations, just logical, plausible, sufficient, coherent, and (hopefully) parsimonious ones.

In particular, I think it’s interesting to note that none of the legal or scientific standards of evidence requires 100% elimination of every other possibility every time in order to conclude that a person is guilty, that an historical event happened, or that an hypothesis is supported or refuted. On the other hand, the Law of Noncontradiction allows us to identify logical options and eliminate directly contradictory statements as equally true, so we can organize the logical outcomes in certain situations.

Extraordinary Evidence?

Finally, let me address the famous objection put forward by atheist astronomer Dr. Carl Sagan: “Extraordinary claims require extraordinary evidence.”[10] I have two responses to this. First, I am interested in evaluating extraordinary claims, such as “God exists” using evidence that is available. And the evidence one requires must be appropriate for the question one has. As I discussed earlier, as a political scientist my research questions drive my research methodology, and the same principle ought to to the question of God’s existence. It seems to me that what Sagan sought was the kind of scientific evidence that is simply inapplicable to the question with which I’m dealing. But my second thought is that I am willing to apply the principle within analytically-appropriate knowledge domains. So I think it is reasonable to expect extraordinary evidence to support Dr. Sagan’s claim that “The cosmos is all that is or ever was or ever will be.”[11] Or that nonliving matter spontaneously became living. Or that the universe started all by itself, creating everything from nothing, for no reason, or that it has eternally existed, or that the multiverse is actually scientifically provable.[12]

In conclusion, in seeking to answer the question Does God Exist?, there are several standards of evidence that come into play to some degree. Few, if any, of them hold absolute proof up as even possible to achieve. So whenever I think it’s possible to do, I will lay out my best understanding of the logical options for an issue. (For example, in the next section I’ll point out that on the question of God’s existence there are only two possibilities: Either God Exists or God Does Not Exist; there is no in-between state.) So there is a true answer! Then, thinking through other related questions, the various standards of evidence will come into play in ways that seem appropriate to the particular aspect of the questions at hand.

Does God Exist?

In this post I address a relatively simple yet profound truth: God either exists or God does not exist. There is no in-between state. What we believe about God’s existence is entirely irrelevant to whether God exists. And whether we can even know about God’s existence is also irrelevant. The fact of the matter is, there is a yes or no answer to this question. The real question is whether we can acquire sufficient reasons to conclude that one of these two is more likely to be true than the other.

Yes or No

Does God exist? There are only two answers, Yes and No, and they both can’t be true. That is, either God Exists or God Does Not Exist. Therefore, I have to evaluate which one of those answers corresponds to reality and the way things actually are, can be logically supported, and is evidentially supported.

Let’s lay out the logical options as a big picture roadmap for my analysis.

  1. One of these two statements is true, and the other is false: Either God Exists or God Does Not Exist.
  2. The question of God’s actual existence is independent of my belief about this. That is, it doesn’t matter whether I believe it or not, God either exists or does not exist.

Therefore, any person’s belief that God does not exist has no bearing on whether God actually exists. Similarly, any person’s belief that God does exist also has no bearing on whether God actually exists.

After I made these points at a recent talk, a young man suggested that there might be a third option—that we can’t know whether God exists. He was confusing knowledge about God with God’s actual existence. (In philosophical terms, the question of God’s reality is ontology, but the question of acquiring knowledge about reality is epistemology, and this was the difference.) Thus, it’s useful to address a third point related to knowing whether God exists:

  1. God’s actual existence does not rely on my possessing knowledge about God or about God’s existence.

What this means is that God may exist even if I don’t know it. Not knowing it does not bear on God’s existence.

It also means that God may exist even if I possess incorrect information about God. The relative correctness of my information does not affect God’s actual existence or actual nature, only my understanding of God.

Finally, it’s important to note that if God exists, there may be evidence of God’s existence. God may or may not have intentionally conveyed information to humanity about himself (or herself, or itself). Even if there isn’t evidence we can perceive, that doesn’t affect whether God exists, only that our knowledge about God doesn’t have evidence. On the other hand, we might be able to perceive at least some or all of the evidence for God’s existence.

(There are some important implications for the answer to these questions, and a person who concludes that God exists or doesn’t exist has some more work to do in accounting for the way the world works, or for explaining God more thoroughly. That set of questions is beyond this series, but are on my agenda to address down the road.)

My First Conclusion: God either exists or does not exist. The arguments and evidence must point me to which one of these statements is true, even if I don’t attain 100% certainty.

In my next post, I will address an important aspect of God’s existence. Most human conceptions of deities suggest that there exists a realm or dimension beyond the physical realm. If this realm is nonmaterial, it is one in which nonmaterial beings or deities could exist. Is there evidence that the non-material realm or reality exists in which a being like God could exist?

Next Post: Let’s Get Real(ity).

[1] An irrational person is someone who ignores the evidence and information and then follows a path unlikely to end with her goal being achieved, or that she just wanted to follow despite the evidence. So it would be irrational of me to use materialist methods to study nonmaterial things, because the method would not reasonably be expected to provide evidence about the subject.

[2] All of this begs the question, “How can a mere collection of neurons, blood vessels and electrical impulses actually do this and have any confidence in the result?” Late in my process I discovered the so-called “Argument from Reason,” which I’ll discuss later. I believe this provides a compelling argument for an intelligent, creative, reason-based being behind everything. But at first I just intuitively believed that I had the capacity to reason to a true conclusion.

[3] OED Online. Oxford University Press, accessed February 20, 2014.

[4] Again, this raises the question of how a mere collection of neurons, etc., determine whether a statement is true or false and then trust themselves that they have come up with the right answer.

[5] Thanks especially to a couple of computer programming courses back in the 1980s, and to Dr. Valli Koubi who took the time to walk our International Conflict class systematically through the logic of Expected Utility Theory when I was at the University of Georgia.

[6] See, for example, Elizabeth Shown Mills, Evidence! Citation & Analysis for the Family Historian (Genealogical Publishing Company, 1997).

[7] See, for example, Gujarti, Basic Econometrics (McGraw-Hill, 2003); an accessible introduction is Cuzzort and Vrettos, The Elementary Forms of Statistical Reason (St. Martin’s, 1996).

[8] This is calculated with a statistic called the coefficient of determination, which is the proportion of the variation in the probability—technically the logged odds—of voting explained by the variables.

[9] Horn, Laurence R., “Contradiction”, The Stanford Encyclopedia of Philosophy (Spring 2014 Edition), Edward N. Zalta (ed.), Accessed at http://plato.stanford.edu/entries/contradiction/ August 29, 2014

[10] Sagan, Carl (writer/host) (December 14, 1980). “Encyclopaedia Galactica”. Cosmos: A Personal Voyage. Episode 12. 01:24 minutes in. PBS. Cited in http://en.wikipedia.org/wiki/Carl_Sagan#cite_note-74, Accessed 2 November 2014.

[11] Sagan, Carl, Cosmos (New York: Random House, 1980), 4.

[12] See the excellent discussion of this claim by Rob Lundberg, “Responding to the ‘extraordinary claims require extraordinary evidences’ Objection,” at http://roblundberg.blogspot.com/2014/01/responding-to-extraordinary-claims.html.