The subject of bias in biblical scholarship is poorly understood and often simply bandied about as a blunt instrument. This isn’t a good idea, so consider the below:
What is Bias
Two of the definitions from the OED are worth relaying:
- Tendency to favour or dislike a person or thing, especially as a result of a preconceived opinion; partiality, prejudice. Also: an instance of this; any preference or attitude that affects outlook or behaviour, esp. by inhibiting impartial consideration or judgement.
- Mr. McKerrow’s paper from first to last shows a strong bias in favor of private-ownership.
- A tendency, inclination, or leaning towards a particular characteristic, behaviour, etc.; a propensity.
- E.g., Mr Gladstone said he would leave it to the Committee to consider whether any provision of that nature should be adopted; but at present his bias was to leave it to the good sense of the presiding officer in the case.
Is bias a problem for biblical scholarship?
On the view that the aim of much biblical scholarship is the publishing of research to be commended to any and all readers in the guild or outside, bias of the first definition is potentially problematic. By the definition, it seems very negative, but by the example, it is not so clear. To elaborate, consider the operative descriptors in the definition, “partiality” and “prejudice.” Prejudice is clearly an academic vice, since it represents a conclusion made prior to investigation. Partiality might represent a pre-orientation, but then it could be argued that one’s positive predisposition is simply based on long held, evidence-based convictions, and does not indicate a conclusion. Thus, it isn’t that partiality is per se a problem: for someone can be partial to a school of thought they once determined was accurate, and so, for example, review a new monograph agreeing with the position with a certain partiality. In fact, this could be a description of the example given above of Mr. McKerrow.
However, even if partiality is tamed into something more respectable like this, such partiality—if it is obstinate, or unwilling to re-examine when necessary—would be a problem, since it could easily form the basis of prejudice.
On this first definition of bias, there must be some daylight allowed into examples and descriptions: since no human being is able to pretend as if he or she makes every decision in a vacuum, and is a perpetually blank slate. The fact of the matter is, everyone will close their minds on some big issues in their field, and this will, in a sense, make them partial or impartial to subsequent scholarship.
Again, the problem becomes what such partiality is held to for its own sake, or when new counter-positions are dismissed for bias’s or prejudice’s sake (counter-positions worth addressing anyway—there are always counter-positions). The issue is not that someone defends a position they are partial to in light of research arguing against it (for taking such research seriously into account is the stuff of scholarship). The issue is when someone rejects or ignores such scholarship out of hand. Only then could the maintenance of a partial position be said to drift into prejudice and problematic bias.
The second definition is less-nuanced definition that essentially fits within my exposition of “partiality” above. It must be admitted that a great many things draw a person’s inclinations. Bias of this sort, is problematic if and only if it can be shown that evidence is being rejected, wilfully ignored, or mishandled.
Thus, in both cases (which constitute the uses of the assertion of bias), the evidence of bias is not determined by stated conviction per se, but by demonstrating a certain kind of evidence: wilful ignorance, rejection of, or mishandling of evidence (research or data) that contradicts the position.
Do all scholars have bias?
All scholars have partialities, since all people do. Prior experiences qualify perspectives on later ones; all people experience through the course of their life, therefore, all people accumulate some pre-qualification for later experiences. All scholars thus potentially nurture biases that could impact research.
How can bias be detected?
Bias of the sort discussed above is particularly difficult to detect in biblical studies since in many cases the data are underdetermined, which means that demonstrating wilful disregard for evidence is not always easy to show. It is often up to scholars to argue in such a way so as to constrain the data within a certain framework, such that the evidence is more determinate. However, in many cases it may simply be impossible to show that real problematic bias is at work when a scholar rejects such presentations.
Is consensus a sure guide to detect bias?
If a consensus emerges on a subject, does this help to identify when a person’s maintenance of a position has drifted into problematic bias? If so, it would work this way: scholar “a” holds position “b,” consensus “c” emerges, but scholar “a” maintains position “b,” therefore, bias is at work in the maintenance of position “b.” Is there good reason to think this is accurate?
Consensus is a notoriously fraught concept. Firstly, consensus adherence is not a feature of paradigm-shifting work, which begins with novel proposals against a ruling paradigm (a novel proposal may well be a re-framing of a debate consistent with some prior-held position [almost no positions are new in biblical studies]); consider the work of Thomas Kuhn if this is new or interesting to you (Kuhn, The Structure of Scientific Revolutions). This means that we should expect important scholarship to be at variance with some consensuses, some of the time.
Secondly, a consensus requires serious qualification in order to be consider a likely place to look for knowledge (and consensus can only be a tool for detecting bias if it actually is a likely place to look for knowledge). Such a consensus should be sufficiently large, heterogenous, and uncoerced. See Aviezer Tucker’s article on consensus (Tucker, The Epistemic Significance of Consensus, Inquiry, 2003). While the point in the previous paragraph remains, it might seem that very “good” consensuses of this type more likely than not expose bias in those who reject them.
Assume this is the case, and that we are dealing with a very strong consensus. It would still need to be shown that the subject in which the consensus emerges is the kind of subject that can offer veritable proofs such that new positions are extremely unlikely to emerge, and the rejection of the consensus is tantamount to the wilful disregarding of obvious evidence.
Is history the kind of field that enjoys such consensuses? On atomic facts, yes: many facts accepted into the guild are clear, evidence-based, and conspiracies need not be seriously entertained unless they accumulate real evidence. But on larger, more synthetic hypotheses, it is not clear that such consensuses often emerge. Take, in New Testament studies, the example of the “consensus” of Markan priority as part of a solution to the synoptic problem. This view is fairly large (most biblical scholars hold to it), that set of scholars is diverse, and very likely uncoerced (I’ve never heard of pressures put on a scholar to hold one position over another on this). Yet it doesn’t seem to most to be a binding consensus, and therefore we don’t call proponents of Matthean priority or something else to be problematically biased because they reject the Markan view, so long as they have an argument. The fact of the matter is, in many cases in historical disciplines, such “strong” consensuses that operate at a higher-order level than the level of the atomic fact are constituted of too many potentially underdetermined pieces of data (linguistic, historical, textual, etc.) to be as binding as, say, consensuses in modern experimental science.
It is important to note that the epistemic significance of consensus, to the degree that there is a positive one, must take into consideration the set of scholars that make up the consensus. There is, for example, a serious problem with the set “biblical scholars.” Even “New Testament” scholars. It may be the case that most adherents to a particular view are relatively uninformed, but assume a position is accurate, thus illicitly filling out a “consensus” in the minds of many.
The relevant set should be the set of scholars who have actively researched in, and published in, an area. For example, the “consensus” position among scholars writing on Acts (commentaries, NT introductions, etc.) is that the work was written ~the 80’s AD: but not one major publication actually on the date of Acts (or the dating of the NT) maintains this position. On this subject, the consensus of published research is that Acts is written earlier, while a secondary consensus is emerging (not as a novel position: it has been held by scholars for over 150 years), that Acts was written later than this time (after the 90’s AD, into the second century).
Then within this narrower set of scholars, things like heterogeneity and coercion are worth meditating on. We tend to trust positions that seem diversely held to, and uncoerced: yet detecting social coercion is very hard to do. Trends emerge as much socially as they do evidentially.
Since diversity of opinion should be a rule for scholarship so as to mitigate trend-setting, consensus doesn’t seem like a very precise tool to use to detect bias in those who don’t belong to it, even while it is not at all unimportant.
What about confessions as a guide?
Many publications are published by confessional institutions (journals, book publishers, or seminaries). That is to say, these institutions have stated commitments to a series of propositions, such that membership in them presumes adherence to the propositions (and in cases, proceeds on contractual agreement).
Confessions seem like a guide to bias, since in these cases prior commitments can be proven. Are confessions of faith a likelier place to look for bias than dissenters from consensus?
I have already shown that all scholars already have prior commitments, so it cannot be the fact of commitments themselves. The question would be ability to modulate such commitments in light of evidence, and herein lies the apparent potential of “confessions” as a tool: it can be shown that in cases where a scholar is beholden to confessions, that scholar may be motivated to maintain commitments financially, or socially, rather than evidentially. So, does this mean such scholars are inherently biased?
It of course cannot mean they are inherently biased, since they may be scholars in a field (an engineer at a Christian university) that has little direct connection to the substance of the confession. But we are discussing biblical scholars. Does research in biblical studies always have a determining relationship with the substance of confessions?
This question of confessional commitment and research instance needs to be nuanced in order to ascertain bias. For example, the drawing of the inference from “God exists” to the conclusion “I favour this textual variant” is not apparently logical without explicit evidence forming premises that make the case. In many cases—even in biblical studies—it will simply be a non-sequitur to claim that some confessional position is a premise in an argument that has as its conclusion the research position of “this” paper or “that” monograph, or, that the conclusion of such a paper is a premise in an argument that has as its conclusion the already-held confessional proposition. What can be shown in these cases is a lack of inconsistency with those positions (in much the same way as the engineer example above): that is to say, there is no direct bearing from the one to the other. In many cases, there may be a consistency between one argument and a confessional element, but even here, the presence of consistency is not at all determinative on its own of bias.
Foe example, consider confessions that state something like “the Bible is the true word of God.” This must surely mean that such biblical scholarship done under such a confession is inherently biased. But consider the kind of scholarship that actually impinges directly on this statement: how often is the conclusion stated or implied from research that the Bible is “therefore not the word of God”? It is a non-sequitur to read a paper on the historical nature of a pericope in the Gospel, and then conclude “therefore the Bible is the word of God”—even if the substance of the paper is consistent with that believe and may in principle play a role in the determination of that belief (being part of a premise in an argument).
Furthermore, the question of what it actually means for people who confess “the Bible is the true word of God” must be raised: since many people see this is a claim to the normative authority of what the Bible teaches about God, while accepting all sorts of positions that would be considered “progressive” within biblical studies as compared to traditional biblical scholarship.
Well, the confession could be more explicit: what if says “the Bible the inerrant word of God”? Even here, the claim in a piece of research, “this text is historically accurate” is too low-definition a claim to be put into the “bias” category even on this confession of inerrancy. This is because (whatever one thinks of this): holding to the literal historical truth of every biblical proposition is not a necessary condition for those confessing inerrancy (I say this sociologically: people who confess inerrancy can demonstrably be shown to re-interpret simple historical claims which makes them historically untrue, but literarily true).
Consider another, potentially more binding example. Consider a scholar, working at a Christian institution, who argues for the historical truth of the resurrection in a monograph. It is easy to see that the conclusion of the research comports with a prior-held belief, and thus that it is unavoidably partial to say the least. In this case then, is problematic bias ever, always, at work?
Even while the terms of the research seem to be coloured with bias, if a scholar simply makes the argument: I am re-examining something to which I already hold, and since I commit myself to some higher order principle (like, that my belief should be evidence based), then I will “follow the evidence where it leads,” then that will likely (or should) assuage concerns. The other way to assuage such concerns is to argue in such a way so as to commend the research on its own, regardless of presumed prior commitments (or even in spite of them).
Now, of course, such a person can truly and honestly make the claim, yet still be beholden to partialities very deep. In this event, it is not adequate to say “bias” against the research, unless that assertion can be maintained by the evidence of the research (namely, that it is wilfully ignorant of, or mishandling, factual, non-underdetermined information). That is the standard.
But what about scholars who claim to maintain their belief in “x” confession on the basis of evidence (or the lack of seeing any evidenced reason to reject it)? The claim that such people are biased is really a claim that they are liars, or at least, are coerced. To assume either is to assume that such scholars care more about a specific job, or specific social network, than their personal moral or intellectual integrity. It is doubtless the case that many do; humans are a vicious lot.
And yet it can be shown that many scholars do leave, change, or lose jobs, or make some other change in order to maintain intellectual / moral integrity. Furthermore, it is obvious that people working in non-confessional places also are motivated to hold certain positions in order to maintain some other kind of status (with a certain publisher, a certain administrator, a certain social group). And in these cases, it may be that confessions clarify what is otherwise obscure: the broad terms of social agreement—much like the conflict-of-interest funding disclaimers on medical research are presumably supposed to work (are there not often undisclosed conflicts of interest?).
The point is, whether we look to consensus, confession, or some other sense or tool we have to detect “bias, we are left with two choices: speculation, or evidence. We should not speculate, and where bias is clear, we won’t have to. Easy dismissals aren’t worth much. In all cases, unless it can be evidentially demonstrated that in “this case” a scholar only maintained “this position” because of a prejudice which led him or her to abuse “this evidence”—then bias is not a worthy term to use in serious academic discussion.
It may be a critical intuition, and you may be right—partiality may even be admitted: but the evidence should be dealt with on its own terms.