How the Document Got Its Authority

Abstract. Purpose: To invite further consideration of and research into the authoritativeness, reliability and trustworthiness of documents. How do documents come to be trusted? Why are some more trusted than others? Approach: The cases of the OED and Wikipedia policies are explored from a historical perspective, and other cases are considered. Findings: Authoritativeness seems inherent to documents because of a cognitive metaphor that says "what is persistent is trustworthy". Practical implications: This feature of documents exposes users to a number of pitfalls related to trusting illegitimate documents. This has important implications for document literacy. Value: New insight into documents is achieved by applying cognitive metaphors and prototype theory to documents.

I swear with my hand on this room's most sacred book, the alphabetized telephone directory, to speak the honest truth.

—Derek Hartfield¹, One and a Half Times Around the Rainbow, 1973

Introduction: The Dictionary and Wikipedia Policy

The Oxford English Dictionary (OED) was conceived in the mid-1800s as a descriptive, historical and exhaustive catalogue of the words in the English language. Descriptive, because it was meant to describe how words were actually used (in print), rather than prescribe how they should be used. Historical, because it sought to record all the ways a given word was used since its entrance into the English language. And exhaustive, because the OED's goal from the outset was to include every last English word, extant and otherwise.

To linguists and lexicographers, the descriptive nature of the OED is perhaps its most important characteristic. And yet, to others---the general public and academics in other disciplines alike---the OED is often invoked as the ultimate authority on correct English. That is, the OED is treated as a prescriptive document. (It seems rather ironic that linguists and lexicographers, in decrying this practice, are essentially prescribing how the OED ought to be used.) Of course, it is entirely natural that the OED come to be considered prescriptive; it is exhaustive and historical, after all. And, it seems, because of these virtues, it has become authoritative. And by authoritative, I mean, if I may be permitted to draw from the OED in this discussion: "proceeding from an official source and requiring compliance or obedience" (Authoritative, adj., 2015). It may even be the case that any dictionary, not just the OED, can't help but be prescriptive. Indeed, we even seem to conceptualize dictionary in general as a singular authority---we say things like, "I'm going to look it up in the dictionary" (rather than a dictionary).

At present I am part of an ongoing research group investigating citation practices on Wikipedia. This online, free encyclopaedia is entirely community-created and -edited. Wikipedia identifies itself as "not a bureaucracy" (Wikipedia, 2015a), and it strives to support an egalitarian democracy of knowledge. It operates according to a set of principles, which have been documented on the "Policies and guidelines" page. As stated in the page's introduction:

Wikipedia policies and guidelines are developed by the community to describe best practices, clarify principles, resolve conflicts, and otherwise further our goal of creating a free, reliable encyclopedia. There is no need to read any policy or guideline pages to start editing. ... Although Wikipedia does not employ hard-and-fast rules, Wikipedia policy and guideline pages describe its principles and best-agreed practices. (Wikipedia, 2015b)

Something interesting has happened here. The policies and guidelines page was originally created to document "habit and consensus" (Wikipedia, 2001). But today there is also an element of authority in these policies and guidelines; their being organized---documented---seems to have allowed for this. Today, material on Wikipedia is regularly changed, removed or challenged if it does not conform to Wikipedia's stated policies. Just as with the OED, the descriptive couldn't help but become prescriptive. Why is this?

The Authority of Documents

Authoritativeness has always been part and parcel to documents. The English word document itself comes from the French document, which in turn came from the Latin documentum, which was derived from the verb docere, ² which is usually translated as "to teach". Lund (2009) argues that the Latin documentum denoted merely an instrument of teaching and that document was not understood as something providing proof until the 17th century. However, I would assert that the notion of proof was bound up in docere even in Antiquity. For proof, as it were, we can look to Cicero's treatise De Oratore, which was written in 55 BCE and discusses, among other topics, the art of speaking persuasively. According to Cicero,

the whole business of speaking rests upon three things for success in persuasion: that we prove what we maintain to be true, that we conciliate those who hear, that we produce in their minds whatever feeling our cause may require. (2:115, emphasis mine)

For prove here, Cicero uses probare. Shortly after, Cicero reiterates these three keys to persuasion, now saying: "these three objects which alone have power to persuade, namely, that the minds of the audience be conciliated, informed, and moved" (2:121, emphasis mine). He has changed the order, but he has also changed the verb; now instead of probare, Cicero uses docere. In the ensuing discussion Cicero uses probare and docere interchangeably. From this we can understand that Cicero's (perhaps typical of the period) notion of teaching amounted to providing proof. In any case, the idea of providing proof has, for at least the past several centuries, been one of the document's most defining characteristics (Buckland, 1997; Lund, 2009). Proof entails the provision of sufficient evidence for the truth of a proposition; the notion of proof thus implies the existence of truth. In this sense, any proof is an appeal to the authority of the evidence at hand. All this means that, if documents are to constitute proof, they must also be authoritative.

With this history and the above cases in mind, I speculate that any given document gains authority simply by virtue of being a document (or being perceived as a document). More formally: There exists in our culture a conceptual category of document, and a characteristic of this category is that items within it have authority. People seem to apply this reasoning to new items in a nomothetic way---through the analogical stories described by Frohmann (2009)---such that things that seem like documents are understood to have the all characteristics of documents, including authority.

But where did this authority come from in the first place? Perhaps it came from metaphor. Lakoff and Johnson (1980) argue that conceptual metaphors structure our perceptions and understanding. For example, they suggest that, at least in our culture, we have a metaphor that says time is money. This metaphor is evidenced by the range of expressions such as "investing", "setting aside", "saving" and "wasting" time, and it influences how we go about "spending" our time. I would suggest that we also have a metaphor that says what is persistent is trustworthy. This metaphor hasn't been manifest as pithily or profusely in our vernacular as the time-is-money metaphor has, but there are still examples to be found. In the Bible, for example, we find counsel to "Trust in the Lord forever, for in God the Lord, we have an everlasting rock" (Isaiah 26:4). In everyday speech, we talk about "building trust", and buildings are persistent. We sometimes characterize our trust as "enduring" and "unwavering". Conversely, we think of untrustworthy people as "a broken reed", "unstable as water", "fast and loose" or "shifty". And fundamentally, the very notion of trust seems to inhere persistence. In his account of human psychological development, Erikson (1950) describes trust as the first psychosocial concept that a child encounters, in infancy. He defines "trust" as believing that a person will do what is expected; in other words, there is persistence between expectation and outcome.

Trust and authority are, of course, not identical, but they are certainly linked. Lankes (2008) describes authoritativeness as the perception of trust. When we trust something, we imbue it with authority. In the most recent annual meeting of the Document Academy, Andreas Vårheim presented on the community-anchoring role of libraries in Japan amidst the 2011 Tohoku earthquake and tsunami. A discussion on why libraries are trusted ensued, during which Niels Windfield Lund observed that this was "because they store documents". The notion of trust also appears in Buckland's (2015) listing of factors determining whether a document is suitable for use. By "trust", he means: "Are we confident enough of the origin, lineage, version and error rate?" (p. 8). In other words: Is it authoritative?

Many documents also seem to gain authority through another means: by being manifest visually. In her discussion of visual forms of knowledge representation, Drucker (2014) argues that visual information---by which she includes everything from codex books to graphical user interfaces---always and inherently presents arguments for a given interpretation, but by its very formulation in a visual medium it obfuscates its interpretive nature, masquerading as mere presentation. Returning to the discussion of metaphor, Lakoff and Johnson (1980) point out that we also have a metaphor that says understanding is seeing ("I see what you mean"); perhaps the "seeing" also facilitates understanding. For us today, the canonical document appears in textual, physical form, which we perceive visually. It may be that the visual formulation of the document aids in making it authoritative. Moreover, Mirzoeff (2015) argues that our information exchange is increasingly occurring in nonverbal, visual forms, such as videos and images; if this is the case, and if visuality contributes to document authority, then visual culture is an important avenue for research into documentation. A corollary of this argument implies that non-visual forms of information---such as orality---are inherently less authoritative than visual ones. Further studies should investigate this. However, it first must be understood that many oral documents also contain elements of visuality, such as the speaker's gestures and facial expressions. This being the case, it might very well be that audio-only documents are seen as less authoritative than in-person, aural-plus-visual documents.

In short, I propose that we inherently understand that things that are persistent are trustworthy. Documents are persistent, and they also have a long cultural history as sources of authority and trust. Documents today are predominantly visual (I consider text a mode of visuality); this seems to aid their authoritativeness. When we encounter things that seem like documents---in that they exhibit some characteristics of things we already understand to be documents---we are likely to ascribe authority to those things.

Pitfalls

Like all generalizations, the phenomenon of nomothetic document authority can be exploited. The Onion, for example, is a satirical news organization that touts itself as "America's finest news source". The Onion began circulating in 1988 at the University of Wisconsin--Madison, and in the ensuing decades it grew into a full-fledged newspaper that was available freely in several U.S. and Canadian cities, and worldwide via subscription. Since 2013, it is an online-only publication. As satire, The Onion draws on the format of traditional newspapers in order to generate humour. The crux of its humour is in The Onion's similarity to real, authoritative newspapers; its stories are funny precisely because, though they are framed as documents, their headlines cannot possibly be taken seriously. Still, occasionally people not privy to the organization will mistake its headlines as trustworthy---again, because of how closely The Onion mirrors the familiar newspaper format.

But not all exploitations are so innocuous. In his book Trust Me I'm Lying, marketer and media strategist Ryan Holiday (2013) discusses the media-manipulation tactics he used on behalf of his clients. His overarching argument is that blog content informs the content of mainstream news sources, and that blog content is easily manipulated. One tactic he reported using to make blog posts appear more authoritative was hyperlinking certain phrases to other, only tenuously related pages. In so doing, he took advantage of a documentary convention of web writing where hyperlinking is used as a way to cite sources. Holiday argues that most readers do not trace these links; rather, they find their mere presence (or implied presence) enough to trust the article's contents. This type of hyperlinking is analogous to citing references in scholarly works, and similar exploitations of documental authority can be found in academia. Bornmann and Daniel (2008), for example, reviewed the literature on citation behaviour and found that, in certain cases, scientists cite documents that they believe readers will find authoritative, in attempt to bolster the authority of their own findings.

In much the same way, we can fall prey to believing in the legitimacy of doctored photographs. Because we intuitively believe that photographs represent objective realities---mistakenly, as Barthes (1977) points out---we assume that all things that seem like photographs also represent reality. Though it is common knowledge that photographs in the media are touched-up (and sometimes heavily altered), it still seems to be the case that we viscerally, subconsciously, interpret these photographs as real. This interpretation may evaporate as soon as our cognition kicks in, but it leaves an emotional imprint. For example, even though we "know" that magazine models don't really look like their pictures, media-influenced body dissatisfaction is still a widespread phenomenon, especially among adolescents. Encouragingly, McLean, Paxton and Wertheim (2013) suggest that media literacy education may curb body dissatisfaction; this, in tandem with our wider discussion here, points to the importance of document literacy in general.

Why does this happen? Cialdini (2009) suggests in his theory of influence that one of the key factors in influence---that is, causing particular thoughts or behaviours in others---is authority. Cialdini has found that people tend to obey figures of authority simply because they are figures of authority, even when the wishes of those authority figures are incompatible with the person's own wishes. Cialdini\'s key assertion is that, when faced with overabundant information, people use generalizations to make decisions. As such, we might posit that Cialdini's observation regarding authority figures can be extended to other sources of authority, such as documents.

Non-Authoritative Documents

Of course, not all documents are authoritative. A tattoo on the forehead of an American college student that says "I'm 21" is certainly documentary in some respects. Being a tattoo, it is certainly persistent and visual. And yet, the bouncer at the door of a bar is unlikely to find the tattoo an authoritative, trustworthy, reliable testament to the student's age. Similarly, a political cartoon may not hold the weight of an erudite essay, and nor might an email message compared to a notarized letter.

Prototype theory can help us make sense of this. First proposed by Rosch in the 1970s and developed further by Lakoff (1987), prototype theory presents a graded view of categorization. Rather than understanding categories as sets of items that all share all of the same features, we can understand that some items in a category are central, or prototypical, and others are peripheral. Central members of the category possess most of the features---or the most important features---of the category, while the peripheral members may have very few of these features. Prototype theory is often exemplified by the category of "bird", wherein "robin" is a prototypical example and "penguin" is a peripheral example. In this way, we should not expect to find all documents to be equally authoritative. Rather, I'd assert that authoritativeness is a feature of documents as a category, and thus prototypical documents are more likely to exhibit authoritativeness.

Conclusion

This has only been a brief foray into the nomothetic authoritativeness of documents, and its chief purpose is to invite further consideration. Future research could help us better understand how documents come to be authoritative---and trusted---which is related to information literacy and seeking. What sorts of documents are authoritative? What sorts of documents are not? If persistence and visuality really are key contributors to document authority, then what about digital documents, which are sometimes less persistent, and oral documents, which may not be visual? Lankes (2008) discusses a shift in the assessment of credibility in digital documents from looking for authority to looking for reliability; is this merely a reflection of the availability of more sources, or does it have to do with the material qualities of digital documents themselves?

Notes

References

"Authoritative, adj", [Def. 1]. (2015, June). In OED Online. Retrieved from http://www.oed.com

Barthes, R. (1977). "The Photographic Message", In S. Heath (Ed. & trans.), Image Music Text (pp. 15--31). London: Fontana Press.

Bornmann, L., & Daniel, H.--D. (2008). "What Do Citation Counts Measure? A Review of Studies on Citing Behavior", Journal of Documentation, 64(1), 45--80.

Buckland, M. K. (2015). "Document Theory: An Introduction", In M. Willer, A. J. Gilliland, & M. Tomić (Eds.), Records, Archives and Memory: Selected Papers from the Conference and School on Records, Archives and Memory Studies (pp. 223--237). Zadar, Croatia: University of Zadar. Available at http://people.ischool.berkeley.edu/\~buckland/zadardoctheory.pdf

Cialdini, R. (2009). Influence: Science and Practice. Boston, MA: Pearson Education.

Drucker, J. (2014). Graphesis: Visual Forms of Knowledge Production. Cambridge, MA: Harvard University Press.

Erikson, E. H. (1950). Childhood and Society. New York: Norton.

Holiday, R. (2013) Trust me, I'm Lying: Confessions of a Media Manipulator. New York: Portfolio.

Lakoff, G. (1987). Women, Fire, and Dangerous Things: What Categories Reveal About the Mind. Chicago: University of Chicago Press.

Lakoff, G., and Johnson, M. (1980). Metaphors We Live By. Chicago: University of Chicago Press.

Lankes, D. (2008). "Credibility on the internet: shifting from authority to reliability", Journal of Documentation, Vol. 64 No. 5, pp. 667--686.

McLean, S. A., Paxton, S. J., Wertheim, E. H. (2013). "Mediators of the Relationship Between Media Literacy and Body Dissatisfaction in Early Adolescent Girls: Implications for Prevention", Body Image, 10(3), 282--289.

Mirzoeff, N. (2015). How to See the World. London: Pelican Books.

Wikipedia. (2001). "Wikipedia:Policies and guidelines", available at: https://en.wikipedia.org/w/index.php?title=Wikipedia:Policies_and_guidelines&oldid=294531

Wikipedia. (2015a). "Wikipedia:What Wikipedia is not", available at: [https://en.wikipedia.org/w/index.php?title=Wikipedia:What_Wikipedia_is_not&oldid=679225040]{.underline}

Wikipedia. (2015b). "Wikipedia:Policies and guidelines", available at: [https://en.wikipedia.org/w/index.php?title=Wikipedia:Policies_and_guidelines&oldid=676277670]{.underline}