themselves to doing something; in a declarative,5 they 'make changes in the world'
(Searle, 1999, p. 150) through their utterance, as in the case of utterances that
declare war, dismiss an employee, adjourn a meeting, or bequeath a watch. In each
case, the act is performed through representation and communication. According to
Searle (1983, pp. 166-167), 'we can perform an act by representing ourselves as
performing it', if we make the representation under the right conditions and
communicate it to someone (or perhaps to a storage system that someone can use).6
Like most attempts at categorisation, Searle's taxonomy has been subject to criticism
(Hancher, 1979; Suchman, 1994), but it remains influential among philosophers
and analysts of speech acts.
It is also possible to apply Searle's categories to written records. As a preliminary
to further analysis, it can be suggested that:
Assertive records make statements or assertions; they are representations of
claims that humans, or computers programmed by humans, make about the
world. Such claims may be about past actions ('I assert that I attended the
conference') or past, present, or future situations ('I confirm that the
company has 200 employees'). In practice, phrases such as 'I assert that' and
'I confirm that' are usually omitted. In creating an assertive record, a human
may merely write 'I attended the conference'; a computer may merely generate
the statement 'Number of employees 200'.
Expressive records (such as 'I apologise') do not make claims about the world;
they 'take it for granted, and simply react to it' (Hancher, 1979, p. 3).
Directive records range from those that merely represent the asking of a
question ('Were all customers satisfied?') to those giving commands or orders
('Contact headquarters immediately!').
Commissive records also vary in strength; a record creator may make weak
or generalised commitments ('I will return next month'), utter promises
('I promise that I will return'), or enter into formal contracts enforceable at
law.
A declarative record represents the making of 'changes in the world', because
communication of the record creates a state of affairs that did not obtain previously.
According to American philosopher Nick Fotion (2000, p. 51), declaratives 'bring
about a change in status or condition just in virtue of being uttered successfully'. As
Searle (1979, p. 17) observed, 'if I successfully perform the act of appointing you
chairman, then you are chairman; if I successfully perform the act of declaring a
state of war, then war is on'. He could have added that if I successfully perform the
act of bequeathing you my watch, then my watch is yours; the act creates a new right
of ownership. Declaratives have close parallels to 'dispositive' records as understood
by diplomatists.
However, it can be argued that declaratives are not the only records that operate in
this way. The issuance of commissive records such as contracts and promissory
notes makes 'changes in the world' by creating obligations that previously did not
exist. Some directives also create obligations; summonses, for example, oblige their
recipients to attend a tribunal. Searle (1995, p. 70) referred to the responsibilities
and rights that are created in this way as 'deontic' phenomena, a term he derived
from the work of Finnish philosopher Georg Henrik Von Wright (1951). They are
not limited to those recognised at law or requiring legal formalities; an obligation,
for example, might arise simply from a promise, made by email, to attend a meeting
or visit a sick relative. Obligations, ownerships, permissions, and other deontic
entities are social realities; they exist in virtue of collective acceptance (Searle, 2008,
p. 27), and they remain in existence only for as long as they continue to be accepted
or recognised.
In the words of philosopher Barry Smith (2008, p. 44), deontic entities have 'an
anchorage in the realm of records and representations'. In forming deontic entities,
as Searle (1995, p. 85) noted, 'the characteristic devices used are deeds, bills of
sale, registration papers, wills,' and other types of official document. In Western
societies, issuing such documents counts as conveying properties or creating rights.
Rights can be expected to persist for a period of time; they outlast the moments of
issuance of the documents that create and confer them. But the documents also
endure long after their moments of issuance, and laws or societal conventions
permit them to function as evidence of title; if I have the deed by which you sold me
your house, I can use it to demonstrate that the rights to the house are mine.
This practice can sometimes be taken further, and documents can become
negotiable instruments. In such instances, a transfer of a physical record constitutes
a transfer of the rights that were created when the record was issued. For example,
when a debtor has given a written promise to pay a sum of money to a creditor, a
transfer of the promissory note to a third party will transfer the right to receive the
sum in question (Jordan, Warren, Walt, 2000). Far from being simply
information about a debt or other obligation, such records seem to embody the
obligation itself. Like the prior notion that a promise imposes an obligation on the
promiser, the notion that obligations can be embodied in records is a human
construct, but constructs of this kind help to shape lives, coordinate human
behaviour, and support social and economic well-being.7
Of course, rights and obligations are not restricted to literate peoples, and in any
human society they can - in principle - be generated orally, without creating written
records. The earliest written records of property rights were assertions that a
purchaser had acquired a property that had been conveyed to him by oral or physical
means; they reported a 'change in the world' that had already occurred when the
record was created. Over time, the idea grew up that such records, if created using
exact written forms, would possess evidential weight in the event of legal disputes.
archives in liquid times
98
geoffrey yeo information, records, and the philosophy of speech acts
5 Searle's use of 'declarative' (or 'declaration') in his taxonomy has sometimes caused difficulty. For scholars
in linguistics, the connotations of 'declarative' are often closer to what Searle called an 'assertive'. Searle's
usage also differs sharply from the use of the word 'declarative' in what is called a 'declarative act' or acte
déclaratif in certain civil-law jurisdictions. However, Searle's usage is widely followed in the literature on
speech act philosophy, and I have retained it here.
6 In her critique of my earlier work, Anneli Sundqvist (2011, p. 283) appears to overlook this point, which is
argued more fully in Yeo (2007, p. 338) and (2010, pp. 100-101). Representations are not merely what
Sundqvist, following Marx Wartofsky, calls 'secondary artefacts', which describe or 'reflect' actions that
have occurred at some earlier time; they can also help to constitute an action.
7 Besides undermining the simple identification of records with 'information', the analysis given here negates
the contention by Victoria Lemieux (2014, p. 77) that, because all records are cultural constructs, all are
representations of 'beliefs'. Searle (1979, p. 4) observed that 'a man who states, explains, asserts or claims
expresses [a] belief', and this seems broadly correct; we may accept that creators of 'assertive' records express
beliefs about how things were, are, or will be. However, 'directive', 'commissive', and 'declarative' records
operate differently; they are underpinned by human conventions, but do not openly express or represent
beliefs about the world.
99