Language Evolution: The Windows Approach addresses the question: "How can we unravel the evolution of language, given that there is no direct evidence about it?"
The LINGUIST List is dedicated to providing information on language and language analysis, and to providing the discipline of linguistics with the infrastructure necessary to function in the digital world. LINGUIST is a free resource, run by linguistics students and faculty, and supported primarily by your donations. Please support LINGUIST List during the 2016 Fund Drive.
Date: Mon, 8 Mar 2004 07:21:08 -0800 (PST) From: Magda Dumitru Subject: Time, Tense, and Reference
Jokic, Aleksandar and Quentin Smith, ed. (2003) Time, Tense, and Reference. MIT Press.
Magda Dumitru, The University at Buffalo, SUNY
This is one exciting book; as the editors mention in their "Introduction", it is an attempt to bring together the philosophers of time and the philosophers of language. Contrary to what the back-cover says, most essays were not "written expressly for this book", but, as mentioned by the editors, they are the result of the April 11-13, 1997 'Time, Tense, and Reference' Conference held at the newly created Center for Philosophical Education (CPE) housed in the Philosophy Department at Santa Barbara City College. The book has two parts, I (The Philosophy of Tensed Language), and II (The Metaphysics of Time). Each part is made out of two sections, A and B. This lettering notation appears to be harmless in Part I (section A "The Semantic Content of Tensed Sentences", including chapters 1 and 2, and B " The Cognitive Significance of Tensed Sentences", including chapters 3-7), but turns out to be unfortunate in Part II, where section A ("Tenseless Theories of Time", including chapters 8-10) refers to the 'B-theory' of Time, while section B ("Tensed Theories of Times", including chapters 11-14) refers to the 'A-theory' of time.
Despite occasional typos - 'Cralg' for 'Craig' (p.1, last line from top); 'include' for 'includes' (p.2, line 3 from top); 'has' for 'have' (p.32, line 8 from top); 'is' for 'it' (p.35, line 11 from bottom), 'which which' for 'which' (p.41, line 8 from bottom), etc. ^Ö the reader will be delighted to follow a very dynamic text, where frequent cross-reference is made by many authors to each other's work; this is the reason why, sometimes, we chose to discuss together closely related articles. The reading is eased by the carefully written introductions: one by the editors, one by Mark Richard (to Part I), and one by Jan Faye (to Part II).
CONTENTS and CRITICAL EVALUATION
Chapter 1 "Outline for a Truth-Conditional Semantics for Tense" (Ernest Lepore and Kirk Ludwig)
Lepore and Ludwig offer a truth-theoretical semantics for tense a la Tarski and Davidson. The goal is to have a unified analysis of tenses for state and event verbs: they are all indexed to the time of utterance (for the simple tenses) or to the reference time (for the perfect tenses), meaning that tenses quantify over times. It follows that "a verb's tense restricts what modifiers it can take". This may explain why the simple past may take present adverbials ('now'), but never future or past ones ('tomorrow' and 'yesterday', respectively).
The analysis of tenses proposed by Lepore and Ludwig seems to encounter at least one problem, related to the asymmetry they propose for the treatment of present tense of event verbs, as opposed to the past and future tenses. 'Mary opens the door' is awkward because, the authors say, there is no further temporal specification (such as 'on the 23 of this month') to index it to the time of utterance. Such an explanation rests on the assumption that the present is a simple tense. One may argue, however, that the simple present is ideal for story telling (e.g. while presenting the plot of a thriller), where events are past as of utterance time, yet present as of reference time, and hence that the present may be a 'present in the past'. Moreover, simple present tense sentences may include present adverbials, which index them to a reference time in the past, not to the time of utterance: 'At this time, Mary opens the door'.
Chapter 2 "Tense and Intension" (Nathan Salmon)
Nathan Salmon proposes an enriched Kaplanian theory of reference, by introducing the notion of 'content-base'. The content of a sentence is doubly indexed to the content- base, time being a separate semantic parameter (since sentence and utterance may have different times). While Salmon's proposal is very fitting for explaining the difference between 'Sometimes, the U.S. president is a Republican', and 'Sometimes, the present U.S. president is a Republican', it is not sure whether it is successful in establishing adequate reference for cases of 'no demonstration, no speaker uses' mentioned in King 2001, such as 'That guy who scored one hundred on the exam is a genius'.
Chapter 3 "Objects of Relief" (Mark Richard)
As Evans before him, Richard believes that Kaplan's notion of 'character' cannot straightforwardly determine content. He proposes for each character C to have an associated set C* of mundane principles determining a conceptual role. Apart from slight oversights such as defining 'S' first as a "collection of principles that determine a conceptual role" (p.179) and next as a "sentence playing the conceptual role" (p.180), or referring to 'a property of sleeping' (p.180), where he meant to say 'a property of being sad', Richard's essay may be open to the following criticism: C* is taken to relate to linguistic tense; we believe that it relates merely to actual time.
Chapter 4 "Tensed Second Thoughts" (James Higginbotham)
Higginbotham responds to critiques by Mark Richard (this volume) of his theory of cross-reference between state verbs such as relief, regret, and anticipation, and the tense of the verb in the embedded sentence. He sets out to solve the problem posed by modal statements such as 'My root canal might not have been over now', which differ from 'My root canal might not have been over as of the present moment', since cross-reference only applies to the latter example. He proposes to separate the indexical predicate from the circumstances of deployment of its concept. Higginbotham's demonstration is only sketched here; reference is made to a forthcoming article.
Chapter 5 "Tensed Sentences, Tenseless Truth Conditions, and Tensed Beliefs" (Anthony Brueckner) and Chapter 6 "Need We Posit A-Properties?" (Mark Richard)
Brueckner's argumentation starts from Hugh Mellor's position that tense is transcendentally ideal: something must have a property P of being present, because being present only has behavioral consequences. Brueckner questions this reasoning: being present is different from other types of properties, such as 'being intelligent'; while individuals may be intelligent any time, presentness depends on context, hence it must be a B-series property. Mark Richard wonders whether Brueckner's conclusion is really proving a weakness in Mellor's argumentation. He concludes that it does not, since tense is transcendentally ideal not because it involves 'empty modes of representation', as Brueckner argues, but because it involves 'artifacts of representation'. For Brueckner, context is real; causation depends on it, and so do beliefs. For Richard (and Mellor), nothing is real; truth is relative to a mode of representation.
The obvious question, after reading both Brueckner's and Richard's articles, is whether there is any reality outside the subject. One may argue that there is, and Brueckner has this intuition, unlike Richard (and Mellor). But this does not necessarily mean that Mellor's argumentation is lacking, as Brueckner claims. What it means is that such an argumentation may not be intuitive.
Chapter 7 "Time Plus the Whoosh and Whiz" (Arthur Falk)
Falk argues that the experience of time's passage is grounded not in language (Kaplan, etc.), but in perceptual mechanisms ('flushing' and 'freshening'). Perception is tenseless, since it's processing (the 'flushing') is B- theoretic and subjective. Pastness and futurity are labels for experiences flushed (episodic memory) or to be flushed (agency). The whoosh is "the flushing and freshening of the now" (p.225). Unlike Higginbotham (this volume), Falk doesn't see presentness as a reflexive act of awareness, but as a "viewpoint in perception by virtue of direct attribution" (p.225). Perhaps the most interesting part of his argumentation, from the linguist's perspective, is his discussion of tense and aspect. Falk argues against a unified treatment of tense and aspect (Reichenbach 1947), since for him aspect is not indexical. Instead, he proposes that "tense operations apply globally to sentences that already have aspect" (p.240).
Falk may seem biased in his discussion of tense and aspect, for he gives past tense examples only: the perfective 'had worked' and 'had been working', the habitual 'used to work', the progressive 'was working', and the nonprogressive 'worked'. One may wish for Falk to elaborate more on his idea that tense always applies after aspect does, since it is not clear why languages allow only certain combinations of tense and aspect (the future, for instance, is notorious for his intolerance of many aspectual values, crosslinguistically).
Chapter 8 "Two Versions of the New B-Theory of Language" (L. Nathan Oaklander)
Oaklander proposes a new B-theory of time, according to which facts are tenseless, despite there being tensed language and thought. He corrects an oversight philosophers of time and language are guilty of: not realizing that the meaning of 'meaning' has 4 distinct meanings: intentional, ontological, pragmatic, and linguistic. Beliefs (A-series), encoded as intentional meaning, are different from, yet made true by B-facts; intentional and linguistic meaning are different for tensed and tenseless sentences.
Oaklander claims that neither thought nor language can represent reality, other than symbolically. His position is thus the first to defy the purpose of the present volume, that of bringing together philosophers of time and philosophers of language. Yet this would be a small price to pay, if his argumentation is convincing -- which it is most of the time, except for claims such as the one that 'this utterance' is not indexical, unlike 'the utterance now' (p.297, in response to a remark by Ludlow).
Chapter 9 "Why Tenses Need Real Times" (Robin Le Poidevin)
The essay is an exact opposite to Oaklander's article (this book), maintaining that tenses correspond to real times. After reviewing the main tensed and tenseless arguments for explaining causal relations in a sentence such as 'It has been raining, and as a result, the pitch is now waterlogged', Le Poidevin concludes that the Date Theory only can survive criticism. Moreover, it can survive it only if times are taken to be real, not reduced to their contents (events), or constructed from actual and possible events and relations.
Le Poidevin's essay is based on an elegant dynamic between various theoretical positions. Paradoxically, this is the reason why it may be open to critique: realism is declared to be the winning candidate only because neither reductionism, nor relativism is proven to be correct. Le Poidevin's argumentation is thus not constructed explicitly in support of realism, but against competing alternatives.
Chapter 10 "Real Tenses" (Milos Arsenijevic)
Arsenijevic proposes a tense logic system including both tenses and B-series dates. The reason for having such a system is that "the concept of the in-the-world-inherent modalities requires the flow-of-time assumption" (p.329). Arsenijevic is dismayed at the thought that the world as we know it is completely deterministic, as the classic tenseless theory suggests. Introducing contingency would mean to classify worlds not just along the dimension real- possible, but also along the dimension accessible- inaccessible. This two-dimensional representation, christened 'ontological wheel of fortune', creates the flow-of-time and would "enable God to know which time is the present time by simply being informed about the number of possibilities at any time at which this is greater than 1" (p.346).
Arsenijevic's Cartesian representation is certainly a better way to understand time. In fact, multiplying dimensions to represent reality has been the way to make advances in physics, from Einstein (4 dimensions) to string theory (16 dimensions). However, these theories have not been able to explain how the big bang happened, and how the world will end. If we look at Arsenijevic's 'wheel of fortune' from this perspective, we notice not only that the beginning of the 'wheel' is given by a different rule than for any of the subsequent times, but also that one cannot distinguish the beginning from the end of the 'wheel', since they give the same outcome: one possibility only.
Chapter 11 "Reference to the Past and Future" (Quentin Smith) and Chapter 12 "In Defense of Presentism" (William Lane Craig)
In his essay, Smith argues for a maximalist tensed theory of time (aka the A-B theory), according to which presentness, pastness, and futurity are first-order properties. This is, in his view, the only approach to fit the correspondence theory of truth (defined as a "relation between a relevant proposition and the concrete state of affairs it is about" (p.365). Craig's essay is constructed as a step-by-step critique of Smith's paper.
First, Craig considers that Smith's classification of tense theories as de-dicto and de-re is abusive: "Smith associates tense de-dicto with presentism and tense de-re with his maximalist A-B theory" (p.392). A closer look at Smith's text, however, suggests that he actually means: "The presentist wishes to say that 'only the present exists', in any sense of 'exist'" (p.385). Smith only claims that certain versions of presentism (Prior's and Craig's, among others) are or may appear to be de-dicto, unlike others (Ludlow's, for instance). Craig's reading is also the reason why he believes that Smith confuses 'correspondence' with 'direct reference' (p.395).
Second, Craig takes Smith's argumentation to be "an attack upon the coherence of Prior's tense logic, rather than on presentism per se" (p.393), based on what Le Poidevin (this volume) calls 'the prehistoric era objection', namely that "there were no truths prior to the existence of beings capable of making utterances or having thoughts" (p.310). We leave it to the reader to select between Prior's own forewarning as presented by Craig (p.394) and Le Poidevin's solution to this objection (p.312). If the former is chosen, then Craig has a point; if the second is chosen, then Smith has a point.
Third, Craig legitimately argues that Smith has not proved presentness to be a property. Craig assumes that Smith equates presentness with existence, and, since philosophers such as Kant, Broad and Moore, and Alston have argued against the property status of existence, it would follow that the same philosophers implicitly have argued against the property status of presentness. However, one would need to have Smith's definition of existence, in order to decide that it is completely analogous to that of Kant, Broad and Moore, or Alston. If it is, Craig has a point; if it isn't, then Smith escapes criticism. The latter's proposal, however, to place presentness, pastness, and futurity "in an ontological category by themselves" (p.363) is unclear, and this is why Craig's alternative to conceive of presentness as a "temporal way of existing" (p.402) seems more straightforward.
Finally, Craig's first conclusion that Smith's maximalist approach is incoherent may turn out to be exaggerated; his second conclusion that any A-B theory of time is just a "teratological hybrid" (p.406) is based on refuting Smith's argumentation and therefore cannot be an exhaustive critique of all theories trying to reconcile A- and B- series concepts.
Chapter 13 "Basic Tensed Sentences and Their Analysis" (Michael Tooley)
The question addressed in this essay is whether tensed sentences are truly primitive. Tooley finds that "tensed sentences are, without exception, analyzable, but in such a way that they can be true only in a dynamic world" (p.446). Tensed sentences are based on 'basic tensed sentences' of the form 'an event is tenselessly present at a specific time'. Tooley is no presentist though, for he proposes to break down the concept of 'being present' into the primitives 'temporal priority' and 'actuality'.
Tooley succeeds in preserving a sound B-series ontology, while defining becoming in terms of actuality. His analysis is reminiscent of other attempts at unifying the advantages of A- and B-series theories, such as those by Falk, Oaklander, and Arsenijevic, in this book.
Chapter 14 "Actualism and Presentism" (James E. Tomberlin)
In his essay Tomberlin takes a chance to express his skepticism towards both presentism and actualism. He claims that neither of these ontologic alternatives meets the following two challenges: objectual quantification in complex causal sentences, and valid truth conditions for sentences without de-re individuals (as in the sentence 'Ponce de Leon searched for the fountain of youth'). Tomberlin cites a solution by Fitch, including an adverbial modifier of 'searched for', and obtaining 'Ponce de Leon searched-for-a-unique-fountain-of-youthly', but finds this treatment lacking in a coherent semantics.
Tomberlin's is a well-argumented essay, but one can be more optimistic and imagine that an adequate semantics, like the re-composing strategy proposed in Van Geenhoven 1998, for instance, may succeed in explaining the structure of Fitch's 'adverbial modifiers'.
Evans, G. (1990) Understanding demonstratives. Demonstratives, ed. by P. Yourgrau, pp.71-96. Oxford University Press.
Fitch, G. W. (1996) In defense of Aristotelian actualism. Philosophical Perspectives 10, pp.53-72.
Kaplan, David (1989a) Demonstratives. Themes from Kaplan, ed. by J. Almog, J. Perry, and H. Wettstein, pp.481-563. Oxford University Press.
Kaplan, David (1989b) Afterthoughts. Themes from Kaplan, ed. by J. Almog, J. Perry, and H. Wettstein, pp.565-614. Oxford University Press.
King, Jeffrey C. (2001) Complex Demonstratives: A Quantificational Account. MIT Press.
Ludlow, Peter (1999) Semantics, tense, and time: An Essay in the metaphysics of natural language. MIT Press.
Mellor, Hugh (1998) Transcendental tense. Proceedings of the Aristotelian Society, Supplementary Volume 72, pp.29- 43.
Prior, A.N. (1957) Time and modality. Oxford University Press.
Prior, A.N. (1967) Past, present, and future. Clarendon Press.
Van Geenhoven, Veerle (1998) Semantic incorporation and indefinite descriptions: semantic and syntactic aspects of noun incorporation in West Greenlandic. CSLI.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Magda Dumitru is a graduate student at The University at Buffalo, SUNY, interested in linguistic representations and reference.