The Damage and Repair of Misattributions

Misattributions are unfortunately commonplace. However their prevalence should be not be an excuse to let them go unchallenged. In a public sense, a misattribution can harm the reputation of a speaker as defamation (typically libel in the written form, slander in the spoken). In a political sense, misattributions are used to improve or denigrate organisations, policy proposals, etc. on the basis of an appeal to authority; in this case, the damage isn't limited to an individual, but rather to the capacity of the society as a whole to make rational decisions. Academics typically have very little tolerance for misattributions, seeing it as a type of falsification of data, and all that follows. Perhaps one of the best collections is the highly recommended They Never Said It : A Book of Fake Quotes, Misquotes, and Misleading Attributions, was compiled by Paul F. Boller Jr. (Emeritus Professor of History Texas Christian University, Oklahoma), and John George Jr. (Professor of Political Science and Sociology Central State University) and was published by Oxford University in 1989.

With the advent of social media the capacity to distribute such falsehoods is, of course, amplified. The combination of a general sense of information overload plus an enhanced capacity in distribution provides opportunities for an enticing quote (subject to confirmation bias), added to a pleasing image, along with superficial "voting" methods (such as Facebook's "like" button or Google's "+1") and capacity to "share". As a recent example, a friend added a picture of Albert Einstein with the quote: "If the bee disappeared off the surface of the globe then man would only have four years of life left." Now being slightly familiar with the works Einstein, it just didn't sound like him. Call that a hunch; a subsequent google search reveals over three million references to the quote. But how accurate is it?

Not at all, it turns out. The remark cannot be located at the Albert Einstein digital archive, it does not appear in Wikiquote (a source which seeks primary evidence), and most importantly the investigative site snopes.com, finds that the earliest use of the quote comes from 1994 and suggests that the false attribution was for political purposes. In any case, even if Einstein did make the quote, it does not seem that the human species for die off as a result. Colony collapse is a real issue in North America and Western Europe, but not to extent suggested. Other animals (butterflies, beetles, flies, birds, etc) are also pollinators. The honey-bee isn't even native to the Americas, and flowering plants certainly existed prior to their introduction. These are, of course, examples of secondary (Einstein archive) sources of evidence. I don't have access to the primary material that makes up the Einstein digital archive; it is a secondary source and the people who put it together may be suppressing the information, although the possibility of such an anti-bee conspiracy between The Hebrew University of Jerusalem, the California Institute of Technology and the Princeton University Press is unlikely.

Here's another example; "Religion is regarded by the common people as true, by the wise as false, and by the rulers as useful" (Seneca The Younger). Unsurprisingly political activists who have a prejudice against religion would find this pithy remark endearing, especially as it sets themselves against both the common people and the ruling class, whilst providing the opportunity to describe themselves as "wise". If only their wisdom was sufficient to check whether the quote was true. Although found on numerous websites, there's is no evidence whatsoever that Seneca The Younger ever made the remark. Indeed, it would be quite out of character; Seneca was a Stoic, who believed in an ascetic life and argued instead that "religion does honour to the gods, while superstitions wrongs them" (On Mercy). A similar quote, also without any reference to primary sources, is misattributed to Lucretius some fifty years prior. Both, it seems, actually are derived from a remark by Edward Gibbon, some seventeen hundred years later when referring to the period of Antonines, themselves around one hundred years after Lucretius or Seneca. Ironically, it is perhaps just as well the latter coined the phrase; "Errare humanum est" - to err is human.

Secondary sources of evidence is good, but primary sources are even better. As a recent example which I have direct involvement refers to an alleged quote by Linus Torvalds, the founder, chief architect, and now project coordinator of the Linux kernel, a free software tool that is used on the overwhelming majority the worlds supercomputers and most servers, etc. The claim was the Linus said "The future is open source everything", and it is a claim that was repeated in numerous books. Again, the supposed quote was quite contrary to sourced remarks from the past. With a strong personal commitment to the principles of open-source and Linux, I tried to find the original source of the claim, without success. When I noticed Linus at an appropriate conference I took the opportunity to ask him directly. The quote apparently was wrong.

Of course, many people who engage in a misquotation don't like being asked to reconsider their statement. I recently encountered an individual who, in many other respects, seems quite intelligent. I questioned a quotation on a .signature file he was using that included the quote "When the facts change, I change my mind. What do you do, sir?", attributing it to the economist John Kenneth Galbraith. Despite the quote being a well-intentioned one, I pointed out that there was no evidence that Galbraith ever made this remark and that it is actually a altered attribution (again with no primary evidence) to John Maynard Keynes. The actual unsourced attribution is "When my information changes, I alter my conclusions. What do you do, sir?" and first appears in "A Treatise of Melancholie" (1940) by Timothie Bright. Despite receiving these new facts said individual has not changed their .signature; when apparently the facts change for them, it seems they prefer to hang on to an assertion that is clearly incorrect. Personally, I cannot fathom the sort of thinking that allows such dissonance to persist.

Whether the result of an failure to research or an act of deliberate malice, it is not just false quotes that constitute mis-attributions. Inattention to qualifiers is also important as the fundamental meaning of a person's remarks can be altered. As a simple example, consider the statement: "I do not like fascism": That is clear enough - one does not like a political system which is anti-individualist, anti-democratic, and demands absolute loyalty to a collective ideal, expressed through a vanguard party. Removing the qualifier "not" however changes the meaning significantly: "I do like fascism", it now reads, the exact opposite of the author's original intention. Such is the power of an absolute Boolean qualifier. Again on personal experience, I recently chanced upon a written copy of an address given at a local church last year. A quotation is given there that "a member said at the Forum, that 'there is empirical evidence that as you grow older you become more resistant to change'". Bemused umbrage is taken at this misattribution, because the quotation is from myself. It is paraphrased of course (I used the word "one" instead of "you", for example). But the critical part is that the misattribution minus the very important qualifier that started the sentence, "On aggregate...".

The inclusion of the qualifier "On aggregate" changes the proposition to one which demands empirical social science. Without it, it has universalistic pretensions which are disproven simply by illustrating a handful of counter-examples, which of course the addresser carries out. This is a fairly miserable rhetorical device, a contemptible pars pro toto synecdoche with no quantitative significance at all. Large-scale, cross-cultural, aggregate statistical analysis of personality traits and political opinions is what is needed, not selective examples or absolute claims. The fact that the author chose to make an appeal to authority (argumentum ad verecundiam), from a literature graduate from some forty years ago and former politician currently in an advocacy role, instead of actually carrying out such inquiries themselves, has not gone unnoticed. This example serves to illustrate a development in the argument here, moving from the importance of quoting correctly, and being attentive to qualifiers, to the matter of statistical verification and avoiding the pitfalls of selecting arguments from authority.

Empirical evidence, by the way, for the question at hand is readily available; a very-well known study is a meta-analysis from 152 longitudinal studies that compares the consistency of personality traits over time. (Roberts, Brent W.; DelVecchio, Wendy F., The rank-order consistency of personality traits from childhood to old age: A quantitative review of longitudinal studies, Psychological Bulletin, Vol 126(1), Jan 2000, 3-25), which shows that as one ages personality traits become more stable. Psychologically, on aggregate, older people are somewhat more resistant to change - and the evidence points to this effect being cross-cultural, and independent of gender. This dovetails with well-known research on political attitudes (e.g., Norval D. Glenn, Aging and Conservatism, The ANNALS of the American Academy of Political and Social Science September 1974 vol. 415 no. 1 176-186., Duane F. Alwin, Ronald L. Cohen, Theodore M. Newcome, Political Opinions Over The Lifespan, The University of Wisconsin Press, 1991), which suggest that whilst older people ("in aggregate", if it needs repeating) may indeed liberalise their attitudes, they do not do so at the rate of the rest of society. One only need to investigate the political opinion polls of today to compare voting intention and attitudes towards contentious social policies of today (e.g., marriage equality, recreational drug reform legalisation) to see further confirmation, although of course there are even stronger correlations (e.g., level of education).

Education and qualifications are a good issue to raise in terms of error-checking. In general, a source from a relevant authority with appropriate qualifications will be more trustworthy than one that is not, and the higher the level of education and more appropriate the qualification the greater the level of trustworthiness. Of course this is nothing to do with the qualifications in their own right, but rather the process of reflexive peer review process by which they are acquired. The the careful scrutiny of propositions and assertions from those with expertise in a particular field, encourages the development of accurate knowledge. Peer review itself is, rather deliciously, subject to peer review. The famous examples of Cyril Burt fabricating data to show the inheritance of intelligence, or the Sokal hoax where a physicist successfully submitted a parody article to post-modernist literature journal, do indicate that the peer review process is far from perfect. Recognising the imperfection has led to improvements in the process; blind and double-blind reviews processes, data archives, open reviews, and even post-publication reviews, all contribute to the increasing precision of the process.

Critics of academic qualifications, often made from those who don't have any, seem to be unaware of the rigorous process in their achievement, especially on the postgraduate level. The first years of undergraduate studies are more the result of teaching; at a higher level it is increasingly the result of research. Researchers engage in a savage engagement of hypothesis-making and hypothesis-breaking; they put up a proposition, backed by the best possible evidence and logic, and it is the role of others to utterly tear that proposition apart and destroy it. To mild-mannered academics, that is their version of collegiality, the community engaged in a common purpose; in the evolution of ideas, only the fittest will survive. It is of course, quite possible to engage in such a collegiality in a non-institutional setting (the talk pages of the various mediawiki projects and snopes serve as examples), but to reject the process in favour of some sort of untested auto-didactic is akin using Mr. Sparky in lieu of licensed electricians, or seeking medical supplies and the correct dosage from the suggestions of Dave The Wild-Eyed Hippy rather than the prescription supplied from a General Practitioner to a qualified pharmacist.

By way of summary and conclusion, it the combination of accurately sourced quotations, careful attention to qualifiers, application of quantitative research, and considered judgement of qualifications, that will lead to high quality claims. Deviations from these requirements will inevitably reduce the quality of claim, and indeed, if used to express a critical component of the proposition, fatally damage the quality, from which no level of human will can lead to its recovery. If making an argument, one must check their material, seek primary or good secondary sources, be exacting in their claims, seek the most comprehensive and detailed range of backing material, from the best sources. If one is not prepared to do this, they simply shouldn't circulate propositions, quotations, assertions, etc., as they will be spreading falsehoods, and exposing themselves to ridicule. If they are prepared to make the effort however, every attempt should be carried out to totally break one's own proposition, before submitting it to peer review, who will also try to break it. At some stage one will have to fight the snake. Be prepared for it.