“What would be the most reliable standards for identifying true prophets?” 

As explored in the intro, answering this question is no simple task, but cognitive biases and poor epistemic standards make it even more challenging. This section introduces these pitfalls and provides examples to help readers recognize and mitigate their subtle influence. 

Cognitive Biases

Cognitive biases are mental shortcuts that can undermine rational judgment. These biases are not character flaws; they are natural byproducts of our brain’s efficient ways of processing information. While cognitive biases cannot be entirely eliminated, they can be mitigated through increased awareness (the goal of this section) and critical thinking strategies (discussed later in the book). 

Two key biases relevant to our discussion are confirmation bias and belief perseverance.

Confirmation Bias 

“Confirmation bias is the tendency to search for, interpret, favor, and recall information in a way that confirms or supports one's prior beliefs or values...”  

“The effect is strongest for desired outcomes, for emotionally charged issues, and for deeply entrenched beliefs. Confirmation bias cannot be eliminated entirely, but it can be managed, for example, by education and training in critical thinking skills.” 

Examples of confirmation bias: 

• A fan of a sports team reads an article highlighting the team’s strengths and weaknesses but only remembers the positive points. 

• A person with left-leaning political views primarily consumes media from left-leaning sources, reinforcing their existing beliefs. 

Why is confirmation bias relevant? 

Prophetic validity is often an emotionally charged topic for many. It frequently involves significant personal investments of time and effort, which can increase potential for confirmation bias. In searching for reliable standards to identify true prophets, this bias may lead individuals to favor standards that validate familiar or liked religious figures while excluding unfamiliar or disliked ones.

Such biased outcomes may be undesirable. As noted in the intro, great benefits may follow for those who set aside biases and accurately identify true prophets.

Belief Perseverance 

Belief perseverance is the tendency to maintain initial beliefs even after receiving new information that contradicts or discredits those beliefs.  

“Beliefs can survive potent logical or empirical challenges. They can survive and even be bolstered by evidence that most uncommitted observers would agree logically demands some weakening of such beliefs. They can even survive the total destruction of their original evidential bases.” 

Examples of belief perseverance: 

• A person maintains their belief that the Earth is flat, even after witnessing experiments that demonstrate curvature over distance. 

• A loyal smartphone user insists their brand is superior, even after repeated hardware failures and evidence of other brands’ advantages. 

Why is belief perseverance relevant? 

Belief perseverance can cause individuals to maintain their conviction that someone is a true or false prophet—even when fair and rigorous standards suggest otherwise. In some cases, opposing evidence may paradoxically strengthen entrenched beliefs.

Such biased outcomes may be undesirable. As stated in the intro, great benefits may accrue to those who successfully set aside biases to accurately identify and follow true prophets.

Importance of Cognitive Biases

The cognitive biases described here (far from a complete list) can oppose our search for reliable standards to identify true prophets. These biases can become stronger in emotionally charged topics and often operate without our conscious awareness. Recognizing and addressing them is crucial, especially in issues where accuracy matters.

However, cognitive biases are not the only obstacles to our task. Poor epistemic standards—unreliable methods of seeking truth—also pose significant challenges. The next section explores examples so they can be recognized and avoided.

Poor Epistemic Standards

Poor epistemic standards are unreliable methods for seeking truth. This section will address three key examples:  emotional reasoning, unfalsifiable standards, and the confirmation algorithm.

Emotional Reasoning 

“Emotional reasoning is a cognitive process by which an individual concludes that their emotional reaction proves something is true...”  

Feelings can provide useful information and play a valuable role in decision-making (e.g., fear can motivate one to escape a dangerous situation). However, they are not always reliable for determining truth.

Examples of emotional reasoning:  

• A person feels lucky and concludes that they will win a risky bet at a casino. Unfortunately, feeling lucky has little influence over the outcome of wagers. 

• A person concludes, based on feeling offended by a statement, that the statement is false. However, feelings of offense have little influence over whether a statement is true or false. 

Why is emotional reasoning relevant? 

The topic of prophetic validity often evokes strong feelings (e.g., video of people expressing strong spiritual feelings about the truth of their respective prophets and religions). Some may prematurely conclude that an individual is a true or false prophet based on these feelings alone.  

This outcome may be undesirable. As stated in the intro, significant benefits can arise from using reliable methods to accurately identify true prophets. 

Unfalsifiable Standards 

Falsifiability is the ability of a statement to be found false. According to this principle, any worthy claim should remain open to the possibility of being challenged by new information. One can learn whether a statement is falsifiable by asking, "Would it be possible to find out if this were false?"

Examples of Falsifiable Statements 

• “All octopi are purple.”  (Falsifiable—Finding an octopus of a different color would render the statement false.)

• “No octopi are purple.” (Falsifiable—Finding a purple octopus would render the statement false.)

• “There is a refrigerator in my kitchen.” (Falsifiable—A thorough search of the kitchen revealing no fridge would render the statement false, assuming the fridge is not invisible.) 

In contrast, unfalsifiable statements can never be shown to be false—no matter how unlikely they seem. 

Examples of Unfalsifiable  Statements 

• “There is an undetectable elephant in my house.”  (Unfalsifiable—Even an exhaustive search could never rule out the presence of an undetectable entity.)  

• “All people have bright halos, only observable by good people.”  (Unfalsifiable—If a halo was not observed, it could always be blamed on the observer not being good enough. Even if every person in the world denied seeing halos, it could be claimed that nobody in the world was good enough to see one.)

Why are unfalsifiable standards relevant? 

Unfalsifiable statements can be useful in many contexts (e.g., expressing opinion) but are unreliable for determining truth or distinguishing between entities (e.g., whether someone is or is not a true prophet). For instance: 

• “All true prophets have a certain glow about them, only observable by the wise.”  (Unfalsifiable—If a "certain glow” was not observed, it could always be blamed on the observer not being wise enough. By this standard, no possible observation or event could categorize someone as a non-true prophet. Therefore, this standard cannot reliably tell whether someone is or is not a true prophet.)

Since unfalsifiable standards cannot be found false, they cannot reliably identify false prophets, nor distinguish between prophets and non-prophets. Therefore, reliance on such standards risks misidentifying non-prophets or even false prophets as true prophets. 

This may be an undesirable outcome. As stated in the intro, great benefits may accrue to those who accurately identify and follow true prophets.

The Confirmation Algorithm  

One extreme application of the confirmation bias is the “confirmation algorithm”—a process by which all observations are unconditionally interpreted to support a predetermined conclusion. Following is an example flowchart of the confirmation algorithm being used to evaluate whether a religious leader is a true prophet.  

Although the confirmation algorithm was used in this example to support a prophetic claim, it can be similarly used to justify any claim (e.g., flat earth, conspiracy theories, cults, scams).  (For similar diagrams, see "Hermetically Sealed Systems of Thought" in the appendix.)

Why is the confirmation algorithm relevant? 

The problem with the confirmation algorithm is that all possible observations are interpreted in a way that reinforces the claim—even observations that would normally weaken or falsify the claim. Thus, the claim cannot be weakened or falsified, and one’s ability to distinguish false from true claims is greatly diminished. 

This may be an undesirable outcome. As stated in the intro, great benefits may accrue to those who accurately identify and follow true prophets.

The confirmation algorithm seems so flawed. Why would anyone use it?

Subtle versions of the confirmation algorithm are common, particularly in emotionally charged topics. Because its operation can be difficult to recognize, three real-world examples are provided: 

Examples of the Confirmation Algorithm—Jehovah's Witnesses 

The following excerpt illustrates how a leader of the Jehovah’s Witnesses responded to criticism by subtly encouraging reliance on the confirmation algorithm: 

“Every person in this organization—every one of Jehovah’s Witnesses including the governing body—we’re all imperfect and thus we are subject to making mistakes. Which means that if you look hard enough, all you will see instead of seeing these amazing things that Jehovah is accomplishing and our tremendous privilege of being part of it, all we are going to see is the errors of men. But when we focus on the good, on the guidance and loving support that we’ve received, aren’t we moved with appreciation to cleave to this body of fellow worshipers? And just look at what Jehovah is accomplishing by means of his imperfect servants. There is no other organization on earth providing spiritual food like the one that we belong to. When we look at all of this, how could we possibly let the failings of just a few people undermine this overwhelming proof of Jehovah’s backing?”  

• “Remain Loyal to Jehovah'' Regional Conference 2016 

This reasoning implies that any mistakes by Jehovah’s Witness leaders were merely “errors of men,” inherently minor, and irrelevant to their divine authority. Accepting this premise, no matter what a leader says or does—whether making false statements or condoning unethical practices—their authority should remain unquestioned. This is the defining characteristic of the confirmation algorithm: all observations are interpreted to support a predetermined conclusion.

The excerpt also implies that those who raise concerns are simply “looking hard enough” for flaws while ungratefully ignoring “these amazing things that Jehovah is accomplishing.” This reasoning could discourage members from offering valuable feedback—feedback that is essential for identifying areas of improvement within any organization.

This example demonstrates how the confirmation algorithm can be used to deflect any criticism and maintain an invulnerable guise of divine authority. Yet the algorithm becomes even more concerning when its use among other religious movements is considered:

Examples of the Confirmation Algorithm—The Seekers 

Dorothy Martin, leader of the Seekers, predicted that a UFO would rescue the group from a global apocalypse at midnight on December 21, 1954. Members had made significant sacrifices of time, money, and relationships for their belief that Martin was a prophet. With confidence, they gathered at the appointed time. 

Midnight came and went. No UFO arrived, and the world remained intact. Did the Seekers recognize Martin as a false prophet and demand their money back?

No. The group sat for hours in stunned silence, seemingly unable to process what had happened. They were trapped in the confirmation algorithm, such that a blatantly failed prediction was not enough to challenge their belief that Dorothy Martin was a true prophet.

Finally, at 4:45 a.m., Martin announced a new revelation: “The little group, sitting all night long, had spread so much light that God had saved the world from destruction.” Far from abandoning their beliefs, the Seekers doubled down, embarking on an urgent campaign to spread their message as widely as possible. 

This account illustrates not only the sheer power of the confirmation algorithm but also its reliance on emotional investment. A neutral observer might have quickly dismissed Martin as a false prophet, especially after such an unequivocal failure. But for the Seekers, their sacrifices of time, money, and personal relationships had deeply entrenched their commitment to Martin as a prophet. Every act of emotional investment had further solidified their belief as true—even unconditionally true. By the time of the failed prediction, the Seekers were locked into the confirmation algorithm, such that it paradoxically strengthened their devotion, as seen in their redoubled proselytizing efforts.

Through this example, we see the confirmation algorithm’s insidious ability to entrench beliefs. But its potential for harm extends even further.

Examples of the Confirmation Algorithm—The Movement of the Restoration of the Ten Commandments of God 

Leaders of the Movement of the Restoration of the Ten Commandments of God predicted the apocalypse would occur on January 1, 2000. As one member proclaimed, “The world ends next year. There is no time to waste. Some of our leaders talk directly to God. Any minute from now, when the end comes, every believer who will be at an as yet undisclosed spot will be saved.”

When the prediction failed, some members became disillusioned and left the movement. Others, having sold all their possessions and donated the proceeds, demanded compensation. Yet hundreds remained faithful, trapped by the confirmation algorithm. For these members, even a failed prophecy could not be interpreted as evidence against their leaders’ divine authority.

The faithful reconvened on the revised apocalyptic date of March 17, 2000. During the gathering, nearby villagers heard an explosion. The building erupted in flames, killing everyone inside. Investigations later revealed the fire had been deliberately set by the movement’s leaders, with the windows and doors boarded up to prevent escape, and that in the weeks leading up to the tragedy, hundreds of additional members were murdered on church property.

This tragic account underscores the confirmation algorithm’s potential for harm. It creates an environment where people can be easily exploited by leaders who do not have their best interests at heart. By lulling followers into a false sense of certainty, the confirmation algorithm strengthens their belief that they’ve found the ultimate truth—only to leave them vulnerable to manipulation, loss, and, in this case, unimaginable suffering.

In summary, the confirmation algorithm—an epistemic method where all observations are unconditionally interpreted to support a predetermined conclusion—is unreliable, dangerous, and wholly inadequate for our central task of identifying true prophets. Recognizing and avoiding this pitfall is important for anyone seeking truth.