Home » Library » Modern Library » Independent Confirmation and the Historicity of Jesus

Independent Confirmation and the Historicity of Jesus


Jeffery Jay Lowder

When skeptics question the existence of Jesus, they often assume that anyone who accepts the historicity of Jesus must be able to provide extra-Biblical confirmation of his existence. According to this view, the New Testament does not provide prima facie evidence for the historicity of Jesus; independent confirmation is needed.

In my opinion, that view is mistaken. But before I explain why, we need to review some general axioms of historical inquiry. As historian David Hackett Fischer points out, all historical investigations begin with a properly framed question. The historian’s job is to answer such questions with verified empirical statements.

Framing Historical Questions

Questions form the basis of historical inquiry by narrowing the scope of the historian’s research. Writes Fischer, “without questions of some sort, a historian is condemned to wander aimlessly through a dark corridor of learning.”[1] Fischer lists six general principles of question-framing:[2]

(1) The question must be operational (that is, it must be resolvable in empirical terms).

(2) The question should be open-ended (that is, it should dictate the kinds of facts which will serve to solve a problem, without dictating the solution itself), but not wide-open (that is, it should guide the inquiry through masses of information).

(3) The question should be flexible (that is, it should be conceived as an approximation, open to infinite refinement).

(4) The question must be analytical (that is, it must help the historian to break down a problem into its constituent parts, so he can deal with them one at a time).

(5) The question must be both explicit and precise (that is, its assumptions and implications must be spelled out in full detail).

(6) The question must be tested (that is, the historian who asks the question must also be the one to answer it).

Fischer suggests that questions which do not meet these standards are unlikely to yield empirical results.

Verifying Empirical Statements

Fischer describes seven “rules of thumb” that should be considered when verifying historical inferences:[3]

(1) The rule of relevance: historical evidence must be a direct answer to the question asked and not to some other question.

(2) The rule of immediacy: an historian must not merely provide good relevant evidence, but the best relevant evidence.

(3) The rule of affirmation: evidence must always be affirmative. Negative evidence is a contradiction in terms — it is no evidence at all.[4]

(4) The rule of responsibility: the burden of proof, for any historical assertion, always rests upon its author.

(5) The rule of probability: all inferences from empirical evidence are probabilistic. Some historical inferences may be much more probable than others, but they are still probabilistic.

(6) The rule of context: the meaning of any empirical statement depends upon the context from which it is taken.

(7) The rule of precision: an empirical statement must not be more precise than its evidence warrants.

According to Fischer, historians who follow these rules are more likely to verify true empirical statements.

Empirical Verification and Independent Confirmation

An issue that is central to the historicity of Jesus is whether one will require independent confirmation of the Gospels. Certainly most Christians would not insist on such a requirement. Yet most nontheists — including those who accept the historicity of Jesus — believe that the New Testament alone is insufficient to verify the historicity of Jesus. Does the historicity of Jesus require confirmation independent of the New Testament?

Criteria of Independent Confirmation (CIC)

Many evangelical Christians would try to sidestep this entire discussion by arguing that historical texts, like their authors, should be presumed ‘innocent until proven guilty’; thus until someone can prove that the New Testament is unreliable, we should a priori accept its claims. Yet those who accept the empirical claims of a historical text bear the burden of proof just as much as those who assert their falsehood; in the absence of such proof we should suspend judgment. Empirical uncertainty thus forms the middle ground between the claim that empirical claims are certainly true and the claim that empirical claims are certainly false.[5]

But if we must suspend judgment on empirical claims in the absence of evidence, then under what conditions are we justified in passing judgment? We are justified in accepting an empirical claim if there is prima facie evidence for that claim. If, on the other hand, the primary source(s) which make an empirical claim do not provide prima facie evidence; then we are justified in not accepting that claim without independent confirmation.

It would be naive to suggest that the only type of historical evidence is textual evidence. However, because most of the alleged evidence for the historicity of Jesus is textual evidence, I think it would be well worth our time to briefly discus what the prima facie concept means when applied to textual evidence. In so doing, I would like to propose some criteria of independent confirmation (CIC) for empiricial claims made in written documents. For purposes of this discussion, I shall label a source ‘independent’ of another source if and only if it does not rely upon that other source for its information. (For example, on the two-source hypothesis, the gospels of Matthew and Luke would be independent of each other, but dependent upon both Q and the gospel of Mark.)

1. Epistemic probability.[6] What is the epistemic probability of the empirical claim? If the epistemic probability is high, that would lessen the need for independent confirmation. Conversely, if the epistemic probability is low, that would increase the need for independent confirmation.[7]

2. Harmony. The criterion of harmony determines the need for independent confirmation based on a source’s harmony with other sources and itself. If a source makes contradictory assertions regarding a historical detail, it will almost surely need external verification with respect to the detail in question. We may also require independent confirmation “if two persons should tell exactly the same story, using precisely the same phrases and following precisely the same order in narrative details.”[8]

But what about situations in which two or more sources are coherent, yet contradict each other? I think this is a very complex issue and one which I will not attempt to resolve here.

3. Integrity. Did an author have a general reputation for truth-telling? If an author were a known liar, or habitually careless or uncritical, that would clearly increase the need for independent confirmation.[9] However, the converse is not true: just because an author was honest does not mean he was accurate. There may still be the need for independent confirmation.[10]

Evangelical apologists like McDowell are notorious for assuming that eyewitness reports — esp. those allegedly found in the New Testament — must either be truthful or based upon lies. Yet anyone familiar with ongoing, experimental research in psychology on the nature of eyewitness testimony knows that an honest observer can be sincerely wrong. Reflecting on a case in which a Catholic priest had been falsely convicted of armed robbery on the basis of eyewitness testimony–only later to be acquitted thanks to a confession by the actual criminal–renown psychologist Elizabeth Loftus writes, “Implicit in the acceptance of this testimony as solid evidence is the assumption that the human mind is a precise recorder and storer of events. Unfortunately it is not…. To be mistaken about details is not the result of a bad memory, but of the normal functioning of human memory. As we have seen, human remembering does not work like a videotape recorder or a movie camera.”[11] As D.F. Hall points out, “[N]ew, misleading information is not only added to memory, it actually alters the content of what the subject is able to remember.”[12] The upshot is that even honest eyewitnesses who try to be objective may in fact remember misinformation.

The results of experimental research on the nature of eyewitness testimony are especially important when considering the more extravagant empirical claims of the New Testament, yet McDowell makes things easy for himself by assuming that eyewitness testimony should be taken at face value. Factors like the retention interval (how much time was there between the incident and the witness’ recollection of that incident?), focus (what captured the attention of the witness?), and especially post-event information (what post-event information could have supplemented the witness’ memory?) and unconscious transference (confusing a person seen in one situation with a person actually seen in another situation) are, so far as I know, never addressed by McDowell in any of his writings; yet these are precisely the issues which he needs to deal with if we are to accept as factual the ‘eyewitness reports’ given in the New Testament and elsewhere.[13]

4. Partiality. This criterion attempts to assess what impact, if any, an author’s bias would have on the need for independent confirmation. This criterion must be used with care, for it would be historically naive to suggest that a document could be dismissed simply because an author has a bias. (If one could dismiss documents on these grounds, then no Jewish historian’s account of the Holocaust and no Chinese historian’s account of the Cultural Revolution could be trusted, because the emotions surrounding the research run too deeply.) On this matter I must agree with presuppositionalists who rightfully state that all authors are biased.

However, this line of reasoning may only be carried so far. There are situations in which bias may disqualify a document. A forced confession extracted under threats of torture, for example, could hardly be considered reliable. The author’s desire to avoid extreme suffering or even death clearly overrode any academic concerns about producing a reliable document. Likewise, if it could be shown that an author’s bias had falsified previous reports, that should give us pause before accepting other reports from the same author on similar issues.

But if the criterion of partiality can be used to increase the need for independent confirmation, it can also be used to decrease the need for independent confirmation. Scholars who specialize in the study of the New Testament have formulated their own criteria in order to identify ‘authentic’ sayings or deeds. One such criterion is the “criterion of embarrassment,” which asserts that if a tradition ’embarrassed’ the early church, then it is probably authentic for the early church would probably not have had a reason to invent such a tradition. Likewise, one could apply similar reasoning to the criterion of partiality as follows: if an author makes an empirical claim which runs contrary to her bias, especially if the claim is embarrassing to her, this would lessen the need for independent confirmation.

5. Primacy. Is the account a primary or secondary source of information? A primary source may be defined as either “the direct [account] of an eyewitness of events” or “accounts secured by contemporaries from eye-witnesses”; “all other accounts derived from this would be secondary sources.”[14]

6. Immediacy. How contemporary is the report? Dating historical documents is absolutely crucial to the historian’s craft; documents which cannot be dated are “worthless for historical purposes.”[15] The reason dating is so important is because late reports are less likely to be accurate than contemporary reports. “The value of a piece of testimony usually increases in proportion to the nearness in time and space between the witness and the events which he testifies. An eyewitness has a good chance of knowing what happened; a reporter distant from the event by only a few years has a better chance than one separated by a century.”[16]

Sufficient confirmation is acheived when the “independent” source does not itself require its own independent confirmation according to the CIC. Consideration of a single criterion will rarely be sufficient to decide whether to require independent confirmation (which need not be in the form of written documents); all of these criteria should be weighed when making that decision.

Objections to CIC

I can anticipate three objections to the criteria of independent confirmation (CIC) at this point. By addressing these objections, I hope to not only establish the validity of the CIC, but to clarify the CIC as well. The three objections are as follows:

Objection #1: “The CIC would destroy all knowledge of ancient history, for most ancient sources make empirical claims which either lack independent confirmation or are epistemically improbable.” If the CIC are valid, then it would become impossible to know anything about what actually happened, for most historical sources would require independent confirmation according to the CIC. Yet such confirmation does not usually exist, therefore we would not know very much about the past.

Response to Objection #1: Even if one assumes that the CIC would destroy all knowledge of the ancient past, this objection involves a fair amount of question-begging. I freely admit that if the CIC are consistently applied, certain historical claims will become unknowable on the available evidence. For it is quite possible (and in fact likely) that there could be historical claims which are true, which require independent confirmation according to the CIC, and which lack such independent confirmation. But the epistemological situation in which one would be justified in not accepting certain empirical claims on the basis of the CIC does nothing to change the ontological situation that those same claims are nonetheless true. More importantly, if the CIC are valid, then the lost items of historical knowledge were never historical knowledge in the first place.

But in fact I think it is incorrect to assume that the CIC would destroy all or even most knowledge of the ancient past. Most of the empirical claims which we accept about antiquity do not have a low epistemic probability and thus according to the CIC would not have an increased need for independent confirmation.[17]

Objection #2: “The CIC falsely assumes that all writers, the majority of writers, or the important writers would have recorded the event in question.” It is quite possible that other writers did not know of the event, did not feel it necessary to record the event in their writings, or that the writings which contained the independent confirmation are no longer extant.

Response to Objection #2: The CIC does not make any assumptions about what other writers would have recorded if the event in question actually happened; that is not even the focus of the CIC. Rather, what the CIC assert is that one is justified in not accepting certain types of historical claims without independent confirmation. The fact that there may be a good reason why an empirical claim is not confirmed in other sources would not refute the fact that the empirical claim needs independent confirmation, if in fact the empirical claim needs independent confirmation according to the CIC.

Objection #3: “The CIC in general, and the criterion of epistemic probability in particular, assume methodological naturalism.” Influenced by “antisupernaturalist” David Hume, I have a priori ruled out the very possibility of historical evidence for miracles by my insistence that epistemically improbable events require independent confirmation. However, since a supernatural worldview is reasonable, one should at least consider the historical evidence for miracles.

Response to Objection #3: The CIC do not assume methodological naturalism for the simple reason that there is no logical contradiction between the CIC and supernaturalism. Indeed, even the Bible endorses a CIC-type approach under certain conditions (Prov 18:17, Num 35:30, Deut 19:15, and especially the account of Jesus’ appearance to doubting Thomas in John 20:26-29).[18]

Although the criterion of epistemic probability does bear upon the probability of miracles, it is much broader than that. It does not assert that we should require independent confirmation of a miracle claim because of alleged supernatural involvement; rather, the criterion of epistemic probability asserts that we should require independent confirmation before accepting epistemically improbable empirical claims. Numerous empirical claims are epistemically improbable but do not involve a miracle; moreover, the epistemic probability of a miracle claim is higher for a theist than it is for a nontheist.

I therefore conclude that these objections fail and that the CIC are valid.

Application to the Historicity of Jesus

I think all of this has two implications for the historicity of Jesus.

First, one should not define ‘historicity of Jesus,’ as many have done, to mean ‘whether the Christ of the New Testament existed — whether Jesus was born of a virgin, performed miracles, etc.’ This is both misleading and ahistorical. In light of Fischer’s principles of question-framing, it is clear that historical questions should not be framed in such a way as to beg other, equally legitimate historical questions. If one were to equate ‘historicity of Jesus’ with ‘whether the Christ of the New Testament existed’, that would make the question, ‘Did Jesus exist?’, equivalent to the question, ‘Was there a Jesus Christ who is the Son of God?’ But this fails to break the issue into its “constituent parts, so they can be dealt with one at a time.” I therefore suggest that we think of the ‘historicity of Jesus’ as meaning ‘whether the Jesus of the New Testament is based upon a person who actually lived’ and not ‘whether this person did the deeds the New Testament claims he did.’

Second, independent confirmation is not necessary to establish the mere existence of the Jesus of the New Testament. There simply is nothing epistemically improbable about the mere existence of a man named Jesus. (Just because Jesus existed does not mean that he was born of a virgin, that he rose from the dead, etc.) Although a discussion of the New Testament evidence is beyond the scope of this paper, I think that the New Testament does provide prima facie evidence for the historicity of Jesus. It is clear, then, that if we are going to apply to the New Testament “the same sort of criteria as we should apply to other ancient writings containing historical material,”[19] we should not require independent confirmation of the New Testament’s claim that Jesus existed.


[1] David Hackett Fischer, Historians’ Fallacies: Toward a Logic of Historical Thought (New York: Harper, 1970), p. 3.

[2] Ibid., p. 38.

[3] Ibid., pp. 62-63.

[4] Consider the difference between ‘positive atheism’ and ‘negative atheism.’ Remember that ‘positive atheism’ is the positive belief in the non-existence of God, whereas ‘negative atheism’ is simply the lack of belief in God. The negative atheist makes no claims and therefore has nothing to prove; the positive atheist does have the burden of proof by asserting a positive claim.

By analogy, then, we may define ‘negative mythicism’ as the lack of belief in the historicity of Jesus and ‘positive mythicism’ as the positive belief in the non-existence of Jesus. This distinction is necessary because knowing that “Jesus did not exist” is different than not knowing that “Jesus did exist.” Those who make the positive claim that “Jesus never existed” have the burden of proof, just as those who make the positive claim that “Jesus did exist.” Michael Martin’s Negative Evidence Principle (NEP) (Case Against Christianity, pp. 46-47) does not apply here because it makes two unwarranted assumptions: (i) that if Jesus had existed, ancient writers would have been interested in recording details about him; and (ii) that historical evidence for Jesus has not been suppressed, destroyed, or otherwise lost.

[5] Robert M. Price, “Apocryphal Apparitions: 1 Corinthians 15:3-11 as a Post-Pauline InterpolationJournal of Higher Criticism Vol. 2, No. 2, pp. 73-74.

[6] For the purpose of this paper, I shall adopt the analysis proposed by Paul Draper that “Relative to K, p is epistemically more probable than q, where K is an epistemic situation and p and q are propositions, just in case any fully rational person in K would have a higher degree of belief in p than in q.” See Paul Draper, “Pain and Pleasure” in The Evidential Argument from Evil (ed. by Daniel Howard-Snyder, Bloomington, IN: Indiana University Press, 1996, p. 27).

[7] Cohen and Nagel argue that “the corroboration of an event by even a respectable number of independent witnesses will generally be held as of no account if the event is known to be contrary to the verifiable body of knowledge called science.” However, this seems confused. Scientific data is always historical information by the time a scientist evaluates it, if for no other reason than there is a delay — however small – between the time the scientist’s sense organs observe the data and the time the sense-data is received by the brain. The upshot is that if one were to strictly follow Cohen’s and Nagel’s principle, it would not even be theoretically possible to overthrow current scientific theories, for the “verifiable body of knowledge called science” would be a priori immune to contrary findings. See Morris R. Cohen and Ernest Nagel, An Introduction to Logic and Scientific Method (New York: Harcourt, Brace, and Company, 1934, p. 334).

[8] Fischer 1970, pp. 146-47.

[9] Ibid., pp. 80-81.

[10] Allen Johnson, The Historian and Historical Evidence (1926; Port Washington, NY: Kennikat Press, 1965), pp. 34-35 and p. 48.

[11] Elizabeth Loftus, Memory: Surprising New Insights Into How We Remember and Why We Forget (Reading, MA: Addison-Wesley, 1980), pp. 161-62.

[12] D.F. Hall, et al, ‘Postevent information and changes in recollection,’ Eyewitness Testimony: Psychological Perspectives (ed. Wells and Loftus, Cambridge: Cambridge University Press, 1984), pp. 126-127.

[13] See Loftus 1980, p. 168.

[14] Johnson 1965, p. 61.

[15] Ibid., p. 51.

[16] Barzun Jacques and Henry F. Graff, The Modern Researcher (Fourth edition, New York: Harcourt Brace Jovanovich, 1985), p. 167.

[17] At least such empirical claims would not require independent confirmation on the basis of the criterion of epistemic probability.

A similar objection can be gleaned from the writings of acclaimed historian Michael Grant: “The CIC is worthless because ancient writers shared so much material from common sources, and even when such a common source cannot be proved or identified, it may still very often be justifiably suspected” (my paraphrase). This criticism was actually made in response to the criterion of multiple attestation used to determine the authenticity of a saying, but the objection could arguably apply to the CIC as well. Applied to the CIC, however, this argument is not really a criticism of the validity or soundness of the CIC, but rather a statement about the availability of independent sources. As such, this objection becomes virtually identical to the second objection and my response is therefore the same. See Michael Grant, Jesus: An Historian’s Review of the Gospels (New York: Collier Books, 1977), p. 201.

[18] Prov. 18:17: “He who states his case first seems right, until the other comes and examines him”

Num 35.30: “If anyone kills another, the murderer shall be put to death on the evidence of witnesses; but no one shall be put to death on the testimony of a single witness.” (Emphasis mine)

Deut 19.15: “A single witness shall not suffice to convict a person of any crime or wrongdoing in connection with any offense that may be committed. Only on the evidence of two or three witnesses shall a charge be sustained.” (Emphasis mine)

[19] Grant 1977, pp. 199-200.

all rights reserved