Hidden Markov models (HMMs) have been successful for modelling the dynamics of carefully dictated speech, but their performance degrades severely when used to model conversational speech. Since speech is produced by a system of loosely coupled articulators, stochastic models explicitly representing this parallelism may have advantages for automatic speech recognition (ASR), particularly when trying to model the phonological effects inherent in casual spontaneous speech. This paper presents a preliminary feasibility study of one such model class: loosely coupled HMMs. Exact model (...) estimation and decoding is potentially expensive, so possible approximate algorithms are also discussed. Comparison of one particular loosely coupled model on an isolated word task suggests loosely coupled HMMs merit further investigation. An approximate algorithm giving performance which is almost always statistically indistinguishable from the exact algorithm is also identified, making more extensive research computationally feasible. (shrink)

Although the view that punishment is to be justified on utilitarian grounds has obvious appeal, an examination of utilitarianism reveals that, consistently and accurately interpreted, it dictates unjust punishments which are unacceptable to the common moral consciousness. In this rule?utilitarianism is no more satisfactory than is act?utilitarianism. Although the production of the greatest good, or the greatest happiness, of the greatest number is obviously a relevant consideration when determining which punishments may properly be inflicted, the question as to which punishment (...) is just is a distinct and more basic question and one which must be answered before we can determine which punishments are morally permissible. That a retributivist theory, which is a particular application of a general principle of justice, can account more satisfactorily for our notion of justice in punishment is a positive reason in its support. (shrink)

In a recent issue of Analysis I gave a critique of some arguments made by Di Nucci concerning the so-called Simple View – the view that an agent performs an action intentionally only if he intends so to act. In turn Di Nucci offers a reply that concentrates on two points. The first has to do with a group of examples, one having to do with waking a flatmate, and the others with routine actions such as shifting gears while driving. (...) I rejected these examples, arguing that it is not at all obvious that in them the Simple View is violated. Additionally, I pointed out that these are not cases of agents doing something they thought they could not do, and so have no direct bearing on the issues of practical rationality that motivate Bratman's treatment of the Simple View, which Di Nucci is concerned to defend . Di Nucci accepts this second point. He claims, however, that I failed to understand the dialectic of his argument. …. (shrink)

The moral case for experimentation on animals rests both on the goods to be realized, the evils to be avoided thereby, and on the duty to respect persons and to secure them in the enjoyment of their natural moral rights. Some experimentation on animals presents no problems of justification as it involves no harm at all to the animals which are the subject of experiments and is such as to seek to achieve an advance in knowledge. Experiments on non-sentient animals, (...) like those on plant life, may harm the subjects but in ways that in themselves need raise no morally significant issues. Moral issues may arise if the subjects of experiments are members of an endangered species and for like reasons but not simply on account of the harm caused by the experiment to the subjects. Other experimentation on animals is to be justified solely in terms of its benefit to animals, even though it involves harm to the animals upon which experiments are performed. There are many poisons and diseases to which animals and not human beings are exposed which give rise to the need for experiments to determine effective therapies, treatments, or other remedies. Examples include annual rye grass toxicity, foot-and-mouth disease, the many fowl diseases, and so on. Human beings commonly have an interest in successfully experimenting to find effective remedies, either a financial interest as with farm animals, race horses and the like, amusement interest as with pets, or a concern not to have a species or variety become extinct. The moral case for such experiments may be strengthened by reference to such interests if they in turn have a moral basis, but generally it does not depend for its validity on this. Other experimentation that harms the animals involved is to be justified in terms of the benefits to be realized and evils lessened for persons by the knowledge and the use of the knowledge gained, where the knowledge is used to secure the enjoyment of their moral rights by persons. Much experimentation benefits both persons, contributing to the securing of their enjoyment of their moral rights, and animals which do not possess moral rights, and is to be morally justified on both counts. Obviously a great amount of experimentation occurs today which does not admit of justification along any of these lines, and hence, is such that it must be condemned as being morally wrong. (shrink)

The first section restates and elaborates on my argument in "rights," "philosophical quarterly", 1965, Arguing that rights are not explicable as claims, Powers, Expectations, Liberties. Equally, Statements about rights, Often being logically prior to such statements, Are not reducible to such statements. In section two, This claim is supported by reference to distinctions it is vital to draw between rights, Which do not parallel those to be drawn between kinds of duties. We need to distinguish "real" rights which may be (...) partially or completely unfulfillable, "conditional" rights which, If morally unfulfillable, Cease to be rights at all, And "prima facie" rights proper which are always rights, Basic, Ultimate, Rights, Whether or not it is physically or morally possible fully to acknowledge them. In section three, The ultimate basis of rights is examined. It is argued that for any moral right to exist, Some right(s) must be intrinsic to a possessor, And hence based on what he is and what he can become. (shrink)

The concept of prima facie duty is deemed important by ross and the author. The author thinks ross and others have not elucidated the concept and the relation between prima facie and 'absolute' duty. He concludes that "we must explain the obligatoriness of absolute duties in terms of prima facie duties, As being derived from them, And not vice versa, As ross attempted." (staff).

IT IS ARGUED THAT LINGUISTIC ANALYSIS DOES NOT DEAL WITH\nTHE PROBLEM OF UNIVERSALS IN A SATISFACTORY WAY. THE\nCONTRIBUTIONS OF RYLE, WITTGENSTEIN AND PEARS ARE\nCONSIDERED. IT IS HELD THAT THE PROBLEM OF UNIVERSALS IS A\nGENUINE METAPHYSICAL PROBLEM AND DOES NOT ADMIT OF BEING\nDISPOSED OF BY CONCEPTUAL ANALYSIS. MOREOVER, THE FAILURE\nOF ATTEMPTS BY LINGUISTIC ANALYSTS HERE MUST CAST DOUBT ON\nTHE SOUNDNESS OF THEIR BOLD ANTIMETAPHYSICAL CLAIMS. IT IS\nCONCLUDED THAT THE PROBLEM OF UNIVERSALS IS NOT PRIMARILY\nONE OF NAMING, BUT RATHER OF RESEMBLANCES. (STAFF).

The maintenance of genetic integrity is of vital importance to all living organisms. However, DNA – the carrier of genetic information – is continuously subject to damage induced by numerous agents from the environment and endogenous cellular metabolites. To prevent the deleterious consequences of DNA injury, an intricate network of repair systems has evolved. The biological impact of these repair mechanisms is illustrated by a number of genetic diseases that are characterized by a defect in one of the repair machineries (...) and in general predispose individuals to cancer. This article intends to review our current understanding of the complex nucleotide excision repair pathway, a universal repair system with a broad lesion specificity. Emphasis will be on the recent advances in the genetic analysis of this process in mammalian cells. (shrink)

In the area of politics one is not surprised to encounter bias, prejudice, rationalization, special pleading; yet rarely does one encounter these phenomena so often, so blatantly, as in the area relating to equality. Indeed, even among those whose job it is to be impartial, rational, informed, namely, philosophers, one finds less objectivity, less evidence of a genuine search for the truth, and a greater concern to press preconceived views than elsewhere in their work. Among Western philosophers, there is a (...) very widespread commitment to equality as a political value. So deep has been and is the commitment in theory at least, to this value, that concern for knowledge and truth have been subordinated to it by many — as in the refusal to allow or determination to discourage research into racial differences, for fear that the results of such research may lead to a lesser political commitment to equality. Yet, for all this professed concern for equality, a genuine lover of equality could be excused for being perplexed by the attitudes of those who profess such a concern for equality. (shrink)

The three works to be examined here are concerned in their various ways with the rationality of ethics. Baier is concerned almost exclusively with bringing out the rationality of ethics, and in the process develops a new/old ethical theory. Ginsberg's concern with "the rational ethic" is rather subsidiary to his main themes, namely the unsoundness of cultural relativism and the truth concerning the relevance of the findings of sociology and other social sciences to ethics. Mackinnon is largely concerned with the (...) meta-ethical problem of the "different styles of reasoning" in ethics, and hence with the problem of the inter-relation of ethics and metaphysics on various levels, logical, historical and psychological. (shrink)

1. We speak of causes in a large variety of contexts and in respect of a large variety of kinds of effects. For instance, we speak of one billiard ball causing the other to move; of diseases as causing death; of drunkenness as a cause of accidents; of the elasticity of the rubber as causing the balloon not to burst under pressure; of the change in temperature as causing the change in the length of the steel girder; of the lemon's (...) acidity as causing it to taste sour; of the cause of the universe, or of matter; of causes of wars, crises, depressions; of causes of other people's actions, and even of their decisions. In brief, the contexts in which talk about causes occurs are where we seek to explain causally the occurrence of happenings, events, changes, phenomena, the existence of things, the possession of properties, human behavior, states of mind, decisions, etc. Hence a satisfactory elucidation of the concept of cause must cover the variety of causes which are pointed to in causal explanations in these areas. (shrink)

Many, including some celebrated liberal theorists, defend liberty on empirical, prudential, utilitarian grounds such that if practical considerations or changed circumstances were to make intolerance more useful than tolerance, they would be committed to a policy of intolerance. Their theories are therefore liberal only contingently. They cannot be denied the title "liberal," for, apart from historical usage, a theory is liberal if it proceeds on the basis of a high evaluation of liberty whether or not the evaluation rests purely on (...) contingent facts. Other liberals defend liberty for its own sake as well as for its utility, and argue that the sole function of the state consists in preserving or promoting liberty. Other theorists stress liberty as the supreme good but take note of the relevance and importance of other goods, e.g., happiness, justice, well-being, fraternity. They too, because of their stress on liberty, and because they include some of the most celebrated exponents of liberalism, must be included as liberals. (shrink)

The frequently occurring photoreceptor patterns in fish are explained using functional and environmental demands in a geometric model. The shape of the double cone provides a number of constructional properties leading to a limited number of appropriate configurations. The probability of their occurrence is estimated from the degree to which the combination of properties of each configuration meets specific environmental light conditions. A row pattern of merely double cones is especially suitable for vision in a dim homochromatic environment; a triangular (...) pattern is quite appropriate for high resolution and accurate movement detection, whereas the known square pattern has a high adaptive capacity to varying spectral distributions. In this context the transforming capacities of both square and row patterns can be understood. (shrink)