Program / Course 13: Foundations of Apologetics / Lesson 4.3
What you will learn
- Distinguish evidence (what is observed or reported) from inference (what is concluded from it).
- Explain why the same evidence can support more than one conclusion.
- Recognize common pitfalls in moving from evidence to conclusion, including confirmation bias, overreach and mistaking correlation for cause.
- Practice stating the gap between evidence and conclusion and saying how much further support is needed.
The big idea
Evidence does not interpret itself. Between what is observed and what is concluded stands an inference, a step of reasoning that may be good or bad. The same fact can support different conclusions depending on the assumptions brought to it. Careful thinkers separate the two. They state what was actually seen or reported, say what they infer from it, and ask what else would explain the same facts. They also watch for habits of mind, such as favoring what pleases them, that tilt the step from evidence to conclusion.
Key terms
- Evidence
- What is observed, reported or otherwise given before any conclusion is drawn from it.
- Inference
- The step of reasoning from evidence to a conclusion.
- Underdetermination
- The situation in which the same evidence fits more than one explanation, so that the evidence alone does not decide between them.
- Confirmation bias
- The tendency to notice, remember and favor evidence that supports what one already believes, and to overlook evidence against it.
- Correlation
- A regular association between two things, such as two events occurring together.
- Causation
- A relation in which one thing brings about another.
- Overreach
- Drawing a conclusion that is stronger or wider than the evidence supports.
- Base rate
- How common something is in general, before the specific case is considered.
Background
Students of the sciences and of history learn early that facts do not speak for themselves. An astronomer sees a point of light, but the claim that it is a distant star involves a long chain of reasoning. A historian reads a letter, but the claim that it shows a plot involves judgments about the writer and the times. In both cases, one must separate what is seen from what is inferred.
This is just as true in religious argument. A person who reports an unexplained recovery has given evidence. The statement that God healed the person is an inference. Both Christians and critics sometimes blur the line, treating the conclusion as if it were itself observed. Learning to keep the two apart is a mark of careful thought.
Step-by-step explanation
Step 1: Separate the observation from the interpretation
Take a simple case. A man is found standing next to a broken window with a stone in his hand. The evidence is that the man is next to the window and holds a stone. The inference that he broke the window may be reasonable, but it is an inference. He may have picked the stone up afterward, or he may have been trying to catch the person who threw it.
To practice, write the evidence in one list and the conclusions in another. Use plain, neutral language for the first list: what was seen, said or found. Only the second list may contain words such as proves, shows, or must mean.
Step 2: Ask what else could explain it
The same evidence often fits several explanations. This is called underdetermination. The early reports that Jesus’ tomb was empty and that his followers said they had seen him alive are evidence. Believers infer that he rose. Others infer that the disciples were mistaken or deceived, or that legends grew. Each side has to say how well its explanation covers all the evidence, and which assumptions it adds.
A useful exercise is to write at least three possible explanations for any piece of evidence, including ones you reject. If you cannot think of any alternative, the evidence may not be as clear as it seems, or you may not have thought hard enough.
Step 3: Beware of habits of mind
People are not perfect reasoners. Confirmation bias leads people to read evidence in the way they prefer. A believer may notice answered prayers and forget unanswered ones. A skeptic may notice failures of faith and overlook the faithful. Another habit is to see patterns where none exist. Another is to hold on to a view because a great deal has been invested in it.
These habits do not belong to one side. Careful thinkers on every side guard against them by deliberately seeking evidence against their own view, by asking friends who disagree to state the strongest objection, and by writing down in advance what would change their mind.
Step 4: Watch common gaps
Several gaps appear again and again. One is mistaking correlation for cause. If two things happen together, one may cause the other, or a third thing may cause both, or it may be chance. Another is overreach: the evidence supports that something happened, but the conclusion claims it happened for a particular reason. A third is hasty generalization, as when a few cases are taken to represent all. A fourth is ignoring the base rate. If a rare event is reported, one must consider how often false reports occur compared with true ones.
Each gap can be stated as a question. Is there another explanation for the connection? Does the conclusion say more than the evidence does? How many cases have been seen, and are they typical? How common is the event, and how common is a mistaken report of it?
Step 5: Match the confidence to the evidence
The last skill is to state a conclusion with the confidence the evidence allows. Phrases such as it is likely that, the evidence suggests, and it is certain that mark differences in strength. Using a stronger word than the evidence supports is a quiet form of overreach.
A mature thinker is willing to say, the evidence supports this much, and no further, and here is what would help to go further. This is not weakness. It makes it easier for others to evaluate the claim and to say where they differ.
- What is the difference between what was observed and what is inferred from it?
- Can you give an example of the same evidence supporting two different conclusions?
- How does confirmation bias affect religious and non-religious people alike?
- What is the difference between correlation and cause?
- How can you tell if a conclusion goes beyond the evidence?
- What practices help guard against your own bias?
Examples
Evidence and inference in a healing report
A woman reports that her illness disappeared after her church prayed for her. The evidence is that the prayer took place and that the symptoms later disappeared. The inference that the prayer caused the recovery goes beyond the evidence unless other causes, such as natural remission, are considered and ruled out. A fair-minded believer would add that God may work through natural means, and a fair-minded skeptic would admit that the case is not a refutation either. This is a constructed example.
Correlation and cause
A survey finds that people who attend church regularly report being happier. This is a correlation. It may mean that church attendance increases happiness, or that happier people are more likely to attend, or that something else, such as stable family life, affects both. The survey alone cannot decide among these.
Overreach in an argument
A speaker shows that the Gospels were written within a century of the events and concludes that everything they say is true. The evidence about dating supports a modest claim, that the writers had access to early memory. The conclusion about every detail goes further than that. The date is relevant but does not settle the matter.
Different perspectives
Catholic tradition
Catholic teaching on miracles and Marian apparitions requires careful investigation before any event is declared worthy of belief, and the Church states that private revelations do not add to the deposit of faith. The process includes medical examination and the search for natural explanations before any judgment.
Evangelical and Protestant writers
Many Protestants emphasize testing claims by Scripture and caution against taking experiences at face value. The New Testament itself calls believers to test the spirits (1 John 4:1) and to prove all things (1 Thessalonians 5:21).
Eastern Orthodox
Orthodox spiritual writers warn against spiritual deception, called prelest, in which a person takes an unusual experience or inference as a sign of holiness. They teach that discernment requires the guidance of experienced elders and the Church.
Philosophers of science
Many philosophers of science note that observations are theory-laden, meaning that what counts as evidence is shaped by existing ideas, and that evidence usually underdetermines theory. They do not conclude that all views are equal, but that conclusions should be held with attention to alternatives.
Psychologists of reasoning
Research on thinking has shown that confirmation bias and motivated reasoning are common in all groups. Some researchers recommend practices such as considering the opposite and writing down reasons before forming a view.
Why it matters
Almost every dispute about religion turns on inference more than on evidence. Two people can agree about what happened and differ about what it means. A student who can separate the two will argue more clearly, will recognize when a conclusion is stronger than the evidence, and will be harder to fool, whether by an eager believer or an eager skeptic. The same skill is a form of honesty about one’s own mind.
Common misunderstandings
Misunderstanding: Facts speak for themselves.
Correction: Facts are interpreted. Different people, with different assumptions, may draw different conclusions from the same facts.
Misunderstanding: Because bias exists, no one can reason fairly.
Correction: Bias is a risk, not a certainty. Practices such as seeking disconfirming evidence reduce its effect.
Misunderstanding: If an explanation is possible, it is as good as any other.
Correction: Explanations differ in how well they fit the evidence, how simple they are and what they assume. Possible does not mean equally likely.
Misunderstanding: A skeptical conclusion needs no inference.
Correction: Doubt is also a position, and the conclusion that something did not happen or cannot be known is an inference that needs support.
Knowledge check
What is the difference between evidence and inference?
Evidence is what is observed or reported before any conclusion is drawn, such as that a tomb was found empty. Inference is the reasoning from that evidence to a conclusion, such as that the body was moved or that the man rose. Inferences can be good or bad and can be questioned.
What is underdetermination?
It is the situation in which the same evidence fits more than one explanation, so the evidence alone does not decide. The remedy is to list the possible explanations, compare how well each accounts for all the evidence and note the assumptions each requires.
What is confirmation bias?
It is the tendency to notice and remember evidence that supports what one already believes and to overlook what goes against it. It affects people of every view. Seeking evidence against one’s own position is one way to reduce it.
Why is correlation not the same as cause?
Two things that occur together may be linked in other ways: one may cause the other, a third factor may cause both, or the link may be chance. Establishing cause needs additional evidence, such as controlled comparison.
What does it mean to match confidence to evidence?
It means stating a conclusion with the strength the evidence allows, and not stronger. Words such as likely, suggests and certain mark different levels. Overstating confidence is a form of overreach and makes it harder for others to evaluate the claim.
Think it through
- Think of a belief you hold strongly. What would you have to see to change your mind?
- When did you last find that the same fact meant different things to two people?
- What bias are you most prone to, and what habit could counter it?
- Is it harder to separate evidence from inference when the topic matters to you? Why?
Independent assignment
Two lists
In your own notebook, write 500 to 700 words. Choose a news story, a religious report or a historical claim. Make two lists: the evidence stated in neutral language, and the conclusions drawn from it. Then write at least three alternative explanations of the evidence, and say what further evidence would help choose among them. Keep this for yourself.
Further study
- Stanford Encyclopedia of Philosophy, Fallacies Read entries on hasty generalization, false cause and related errors.
- Stanford Encyclopedia of Philosophy, Philosophy of History Read for how historians move from sources to conclusions.
- Stanford Encyclopedia of Philosophy, Religion and Science Read the introduction on how evidence and interpretation interact in science and religion.
- N. T. Wright, The Resurrection of the Son of God See how a historian lays out alternative explanations of the same evidence.
- 1 John 4:1 A call to test the spirits.
- Acts 17:11 Read about the Bereans who searched the Scriptures to see whether the claims were so.