There is now considerable evidence that human sentence processing is expectation based: As people read a sentence, they use their statistical experience with their language to generate predictions about upcoming syntactic structure. This study examines how sentence processing is affected by readers' uncertainty about those expectations. In a self-paced reading study, we use lexical subcategorization distributions to factorially manipulate both the strength of expectations and the uncertainty about them. We compare two types of uncertainty: uncertainty about the verb's complement, reflecting (...) the next prediction step; and uncertainty about the full sentence, reflecting an unbounded number of prediction steps. We find that uncertainty about the full structure, but not about the next step, was a significant predictor of processing difficulty: Greater reduction in uncertainty was correlated with increased reading times. We additionally replicated previously observed effects of expectation violation, orthogonal to the effect of uncertainty. This suggests that both surprisal and uncertainty affect human RTs. We discuss the consequences for theories of sentence comprehension. (shrink)

Across languages of the world, some grammatical patterns have been argued to be more common than expected by chance. These are sometimes referred to as language universals. One such universal is the correlation between constituent order freedom and the presence of a case system in a language. Here, we explore whether this correlation can be explained by a bias to balance production effort and informativity of cues to grammatical function. Two groups of learners were presented with miniature artificial languages containing (...) optional case marking and either flexible or fixed constituent order. Learners of the flexible order language used case marking significantly more often. This result parallels the typological correlation between constituent order flexibility and the presence of case marking in a language and provides a possible explanation for the historical development of Old English to Modern English, from flexible constituent order with case marking to relatively fixed order without case marking. In addition, learners of the flexible order language conditioned case marking on constituent order, using more case marking with the cross-linguistically less frequent order, again mirroring typological data. These results suggest that some cross-linguistic generalizations originate in functionally motivated biases operating during language learning. (shrink)

Recent years have seen a surge in accounts motivated by information theory that consider language production to be partially driven by a preference for communicative efficiency. Evidence from discourse production (i.e., production beyond the sentence level) has been argued to suggest that speakers distribute information across discourse so as to hold the conditional per-word entropy associated with each word constant, which would facilitate efficient information transfer (Genzel & Charniak, 2002). This hypothesis implies that the conditional (contextualized) probabilities of linguistic units (...) affect speakers’ preferences during production. Here, we extend this work in two ways. First, we explore how preceding cues are integrated into contextualized probabilities, a question which so far has received little to no attention. Specifically, we investigate how a cue's maximal informativity about upcoming words (the cue's effectiveness) decays as a function of the cue's recency. Based on properties of linguistic discourses as well as properties of human memory, we analytically derive a model of cue effectiveness decay and evaluate it against cross-linguistic data from 12 languages. Second, we relate the information theoretic accounts of discourse production to well-established mechanistic (activation-based) accounts: We relate contextualized probability distributions over words to their relative activation in a lexical network given preceding discourse. (shrink)

We welcome the proposal to use forward models to understand predictive processes in language processing. However, Pickering & Garrod (P&G) miss the opportunity to provide a strong framework for future work. Forward models need to be pursued in the context of learning. This naturally leads to questions about what prediction error these models aim to minimize.

The existence of monsters throws doubt on life's ability to teach us order. This doubt is immediate, no matter for how long a time we have had confidence, no matter how accustomed we have been to see honeysuckle grow on honeysuckle vines, tadpoles become frogs, mares suckle colts, and in general to see like engender like. It is sufficient that this confidence be shaken once by a morphological variation, by a single equivocal appearance, for a radical fear to possess us. (...) Perhaps, fear, you will say; but why radical? Because we are living beings, real effects of the laws of life, and in our turn future causes of life. A failure on the part of life concerns us doubly: a failure could affect us, and we could cause a failure. It is only because we men are living beings that a morphological failure is, in our eyes, a monster. Suppose we were pure reason, a pure intellectual machine for observing, calculating, and accounting—inert, and indifferent to the objects that give rise to thought: in that case the monster would simply be that which was different from the ordinary, of an order other than the most probable order. (shrink)

Christiansen & Chater propose that language comprehenders must immediately compress perceptual data by “chunking” them into higher-level categories. Effective language understanding, however, requires maintaining perceptual information long enough to integrate it with downstream cues. Indeed, recent results suggest comprehenders do this. Although cognitive systems are undoubtedly limited, frameworks that do not take into account the tasks that these systems evolved to solve risk missing important insights.

The most illustrious tradition of romantic poetry on oriental subjects, from the Westöstlicher Diwan to Rückert, Platen, Hugo and Leconte de Lisle, was inspired essentially by Indian and Persian epic, lyric and gnomic poetry, referring only to a minor degree to the ancient poetry of the Arabs—although it is precisely Rückert to whom we are indebted for a version of the Hamâsa, a famous anthology of pagan Arabic poetry. Goethe approached this anthology through the versions of Jones, described it briefly (...) in the “Notes and Commentary” to the Diwan, and among other sections retold one tale splendidly in verse: the “Song of Vengeance” of Taâbbata Sharran. Here the force of his genius lifted him well above the bristling Latin of Freytag, through whom he had come to know the song, and brought him very close in lyrical intuition to the original of this masterpiece of debated authenticity. But on the whole, the poetic heritage of the ancient Arabs was not adequately echoed in the literature of the romantic age. Rückert himself, inspired by pagan Arabic poetry, is read by few people today; and the poetry that was his model is studied rather by philologists, for its documentary value. Theodor Noeldeke, at the beginning of his classic study of the Mu'allakat, declared frankly: “It is questionable whether the aesthetic pleasure provided by the study of ancient Arabic poetry is worth the great pains required even to approach an understanding of the same.” It is difficult to stand in opposition to the authority of the “old master” of German orientalists, who confessed that “the more he studied the Semites, the more he loved the Greeks.” Even the most humble author of these lines dares to appropriate for himself the lines of Noeldeke; and yet, beyond every comparison, which cannot help resulting unfavorably for the ancient poets of the desert, he who knows how to love and seek out poetry will find in them authentic artistic values. Even the harsh pagan song can in certain passages interest us, fill us with enthusiasm, and move us. (shrink)

The changes in technology applied by industry, whose rhythm has been ever accelerating in the course of the last decades, have destabilized institutions in western societies: they change the meaning of these institutions, and modify their efficacity, without our collective awareness, and without our jettisoning the ballast of inertia and social anachronism. The techniques of industrialization which incomparably endow the great nations and national empires, and which the infant nations desire with a sure instinct, give rise to profound contradictions in (...) the traditional doctrine and practice of national sovereignty. (shrink)

By its method of posing problems successively, the curiosity of modern historians towards antiquity may sometimes give the impression of snobism or of complaisance towards a “fashion,” even when it is actually following a logical bent : just before the last war the multiplication of works on the “imperial cult,” or the “imperial mystique” of the first centuries of our era, presented dangerous temptations for exploitation in interpretations favorable to the rule of personal authority. Notably in Germany, the most serious (...) and objective study of the notion of “principat” among the first Caesars found itself compromised even by the vocabulary with which the Latin words were translated: how many dissertations of articles appeared at that time on ancient “Fuehrertum”! (shrink)

It is evident to anyone who is at all familiar with the documents and witnesses of the times that Aztec society at the time of the discovery of Mexico appears to be intensely religious, that all public and private life was invaded, so to speak, by rites and dominated by beliefs. But what exactly was the place of religion and its ministrants in the hierarchy of powers? To what degree was the priestly function interwoven with that of government and the (...) administration of the city? Can the regime of ancient Mexico be described as a theocracy? To answer these questions, it is of no small use, first, to go back to the period preceding the foundation of Mexico-Tenochtitlan. (shrink)

Essays van acht hedendaagse Nederlandse literaire schrijvers over hun verhouding tot de filosofie.

When, in the year 221 B.C., China found itself unified under the authority of the lord of Ts'in, the latter decided that his title of king (wang) no longer corresponded to his actual glory. After deliberating with his counselors, he decided that from then on his title would be Houang-ti, an expression which we translate as emperor. And if the dynasty founded by Ts'in Che houang-ti was ephemeral, the title Houang-ti was destined to remain that of the Chinese monarchs until (...) the revolution of 1911. This term, formed by two characters, was a neologism; the monarchs of preceding historical dynasties, Yin and Tcheou, were kings (wang). But if the double expression was new, houang and ti were two very ancient words, belonging to religious terminology. (shrink)

We live in a universe infinitely more complex than that which is evoked by the word reality. Only the desire for pragmatic knowledge allows us to believe that things are simply that which they are: the bearers of material qualities by which we distinguish or manipulate them. We give them names, which designate their genre, and make use of them according to our fancy. They are tools or means which refer us to other things to which they have a relation. (...) When knowledge is elevated to a science, in doing away with appearances we discover their structure, and new types of relations, expressed in the language of figures and numbers, beyond which there is only the possibility of other structures and other numbers. The object is explained either by the finiteness of human needs or by the network of scientific relations. It is what it is, nothing more. (shrink)