КАК ВОЗНИКЛО САМОЕ ПОПУЛЯРНОЕ В МИРЕ СЛОВО?
Мэлор СТУРУА
На главную страницу
Темы дня:
• Немецкий ученый нашел золото майя
• Найден простой способ избежать ухудшения памяти
• Одно из главных возражений против дарвиновской теории эволюции, похоже, снято
• Лунная орбита: будет ли остановка?
• Ученые готовятся опровергнуть все, что человечество знает о Вселенной
• Предложено объяснение небывало жаркого российского лета 2010 года
• Генетики получили новые данные о родине человечества
Приближается день рождения слова ОК. Это самое употребляемое восклицание в американском английском - и самое распространенное в мире иностранное выражение.
ОК возникло как хохма. Оно впервые появилось 23 марта 1839 года на страницах газеты "Бостон морнинг пост". Это было время, когда американцы увлекались аббревиатурами и сокращали все и вся. Смыслом шутки была насмешка над неграмотностью любителей сокращений. Слова "all correct" превратились в ОК, потому что люди часто писали "oll korrect".
На мировую орбиту ОК вывел президент США Эндрю Джексон - стараниями своих недоброжелателей. Великий генерал был слаб в грамматике. И поэтому, как утверждали его враги, на документах, которые он одобрял, ставил ОК - от все тех же безграмотных "oll korrect". Но это легенда. Впрочем, она навечно впорхнула в жизнь Америки, как и всего мира. Шутку подхватил телеграф. Телеграфисты стали употреблять ОК как сокращенное от "all clear" ("все ясно"). Первым словом, произнесенным астронавтом Нилом Армстронгом на Луне, было ОК. "Парни, готовы? ОК, поехали!" - воскликнул молодой бизнесмен и преподаватель воскресной школы Тодд Бимер, пассажир самолета, захваченного в трагический день - 11 сентября 2001 года. Вместе с тремя другими смельчаками он вступил в схватку с террористами - благодаря их героизму Боинг-747, взявший курс на Белый дом, не достиг цели, он рухнул неподалеку от Питтсбурга.
Технология дала новый толчок популярности словечка. В программах компьютеров "Apple" родилась клавиша "ОК". ОК говорят люди во всем мире - от таксистов до профессоров словесности. И вот недавно ОК удостоилось чести, которой до него не заслужило ни одно слово в английском лексиконе и, возможно, в мировом. Вышла целая книга, посвященная исключительно ему! Называется она так: "ОК. Невероятная история величайшего слова Америки". Автор - Аллан Меткалф, профессор английского языка и исполнительный секретарь Американского общества диалектов.
Если верить книге, само божественное слово, произнесенное в начале, было ОК. Это, конечно, шутка, принадлежащая мне, но в принципе автор не так уж далек от обожествления ОК. По крайней мере, он считает его "самым великим изобретением Америки".
Конечно, это не самое великое изобретение Америки, но несомненно самое заразное слово на земле, побивающее даже русский мат. В чем же тайна его привлекательности? Все дело в эстетических контрастах, утверждает автор. С одной стороны - идеальный овал О, с другой - конструкция линий-колышков К. Женское и мужское начала. Всего две буквы, но в них зашифрован весь английский алфавит от A до Z. Дело не только в написании, но и звучании. Эти буквы легко и естественно произносятся вместе на любом языке. Именно поэтому ОК стало достоянием всего человечества.
В XX веке ОК стало для иностранцев символом американской простоты, прагматизма и оптимизма, пишет Меткалф. Для современных американцев ОК - "философия из двух слов, отражающая терпимость и восхищение разностью". Наконец, кроме письменного и звукового ОК есть еще и визуальное. Это когда человек делает О, соединяя в круг большой и указательный пальцы и растопыривая остальные - как бы давая понять, что у него все отлично. Пользуются этим жестом чаще всего политики, спортсмены, люди шоу-бизнеса, которым больше и чаще всего приходится доказывать, что у них дела в порядке - то есть ОК.
Оцените этот текст
1
2
3
4
5
пятница, 18 марта 2011 г.
Life’s Irreducible Structure
Michael Polanyi *
Live mechanisms and information in DNA are boundary conditions with a sequence of boundaries above them
Content
Introduction
Living Mechanisms Are Classed with Machines
DNA Information Generates Mechanisms
DNA Acts as a Blueprint
Some Accessory Problems Arise Here
We Can Now Recognize Additional Irreducible Principles
Biological Hierarchies Consist of a Series of Boundary Conditions
Understanding a Hierarchy Needs “from-at” Conceptions
The Sequence of Boundaries Bears on Our Scientific Outlook
Summary
References and Notes
Science, New Series, 160 (3834)
June 21, 1968, 1308-1312.
Introduction
If all men were exterminated, this would not affect the laws of inanimate nature. But the production of machines would stop, and not until men arose again could machines be formed once more. Some animals can produce tools, but only men can construct machines; machines are human artifacts, made of inanimate material.
The Oxford Dictionary describes a machine as “an apparatus for applying mechanical power, consisting of a number of interrelated parts, each having a definite function.” It might be, for example, a machine for sewing or printing. Let us assume that the power driving the machine is built in, and disregard the fact that it has to be renewed from time to time. We can say, then, that the manufacture of a machine consists in cutting suitably shaped parts and fitting them together so that their joint mechanical action should serve a possible human purpose.
The structure of machines and the working of their structure are thus shaped by man, even while their material and the forces that operate them obey the laws of inanimate nature. In constructing a machine and supplying it with power, we harness the laws of nature at work in its material and in its driving force and make them serve our purpose.
This harness is not unbreakable; the structure of the machine, and thus its working, can break down. But this will not affect the forces of inanimate nature on which the operation of the machine relied; it merely releases them from the restriction the machine imposed on them before it broke down.
So the machine as a whole works under the control of two distinct principles. The higher one is the principle of the machine’s design, and this harnesses the lower one, which consists in the physical-chemical processes on which the machine relies. We commonly form such a two-leveled structure in conducting an experiment; but there is a difference between constructing a machine and rigging up an experiment. The experimenter imposes restrictions on nature in order to observe its behavior under these restrictions, while the constructor of a machine restricts nature in order to harness its workings. But we may borrow a term from physics and describe both these useful restrictions of nature as the imposing of boundary conditions on the laws of physics and chemistry.
Let me enlarge on this. I have exemplified two types of boundaries. In the machine our principal interest lay in the effects of the boundary conditions, while in an experimental setting we are interested in the natural processes controlled by the boundaries. There are many common examples of both types of boundaries. When a saucepan bounds a soup that we are cooking, we are interested in the soup; and, likewise, when we observe a reaction in a test tube, we are studying the reaction, not the test tube. The reverse is true for a game of chess. The strategy of the player imposes boundaries on the several moves, which follow the laws of chess, but our interest lies in the boundaries - that is, in the strategy, not in the several moves as exemplifications of the laws. And similarly, when a sculptor shapes a stone or a painter composes a painting, our interest lies in the boundaries imposed on a material, and not in the material itself.
We can distinguish these two types of boundaries by saying that the first represents a test-tube type of boundary whereas the second is of the machine type. By shifting our attention, we may sometimes change a boundary from one type to another.
All communications form a machine type of boundary, and these boundaries form a whole hierarchy of consecutive levels of action. A vocabulary sets boundary conditions on the utterance of the spoken voice; a grammar harnesses words to form sentences, and the sentences are shaped into a text which conveys a communication. At all these stages we are interested in the boundaries imposed by a comprehensive restrictive power, rather than in the principles harnessed by them.
Living Mechanisms Are Classed with Machines
From machines we pass to living beings, by remembering that animals move about mechanically and that they have internal organs which perform functions as parts of a machine do - functions which sustain the life of the organism, much as the proper functioning of parts of a machine keeps the machine going. For centuries past, the workings of life have been likened to the working of machines and physiology has been seeking to interpret the organism as a complex network of mechanisms. Organs are, accordingly, defined by their life-preserving functions.
Any coherent part of the organism is indeed puzzling to physiology - and also meaningless to pathology - until the way it benefits the organism is discovered. And I may add that any description of such a system in terms of its physical-chemical topography is meaningless, except for the fact that the description covertly may recall the system’s physiological interpretation - much as the topography of a machine is meaningless until we guess how the device works, and for what purpose.
In this light the organism is shown to be, like a machine, a system which works according to two different principles: its structure serves as a boundary condition harnessing the physical-chemical processes by which its organs perform their functions. Thus, this system may be called a system under dual control. Morphogenesis, the process by which the structure of living beings develops, can then be likened to the shaping of a machine which will act as a boundary for the laws of inanimate nature. For just as these laws serve the machine, so they serve also the developed organism.
* The author is a former Fellow of Merton College, Oxford, and Emeritus Professor of social studies at the University of Manchester, where he had previously held the Chair of Physical Chemistry. His present address is 22 Upland Park Road, Oxford, England. This article is an expanded version of a paper presented 20 December 1967 at the New York meeting of the AAAS. The first half of the article was anticipated in a paper published in the August 1967 issue of Chemical and Engineering News.
1308
A boundary condition is always extraneous to the process which it delimits. In Galileo’s experiments on balls rolling down a slope, the angle of the slope was not derived from the laws of mechanics, but was chosen by Galileo. And as this choice of slopes was extraneous to the laws of mechanics, so is the shape and manufacture of test tubes extraneous to the laws of chemistry.
The same thing holds for machine-like boundaries; their structure cannot be defined in terms of the laws which they harness. Nor can a vocabulary determine the content of a text, and so on. Therefore, if the structure of living things is a set of boundary conditions, this structure is extraneous to the laws of physics and chemistry which the organism is harnessing. Thus the morphology of living things transcends the laws of physics and chemistry.
DNA Information Generates Mechanisms
But the analogy between machine components and live functioning organs is weakened by the fact that the organs are not shaped artificially as the parts of a machine are. It is an advantage, therefore, to find that the morphogenetic process is explained in principle by the transmission of information stored in DNA, interpreted in this sense by Watson and Crick.
A DNA molecule is said to represent a code - that is, a linear sequence of items, the arrangement of which is the information conveyed by the code. In the case of DNA, each item of the series consists of one out of four alternative organic bases [1]. Such a code will convey the maximum amount of information if the four organic bases have equal probability of forming any particular item of the series. Any difference in the binding of the four alternative bases, whether at the same point of the series or between two points of the series, will cause the information conveyed by the series to fall below the ideal maximum. The information content of DNA is in fact known to be reduced to some extent by redundancy, but I accept here the assumption of Watson and Crick that this redundancy does not prevent DNA from effectively functioning as a code. I accordingly disregard, for the sake of brevity, the redundancy in the DNA code and talk of it as if it were functioning optimally, with all of its alternative basic bindings having the same probability of occurrence.
Let us be clear what would happen in the opposite case. Suppose that the actual structure of a DNA molecule were due to the fact that the bindings of its bases were much stronger than the bindings would be for any other distribution of bases, then such a DNA molecule would have no information content. Its codelike character would be effaced by an overwhelming redundancy.
We may note that such is actually the case for an ordinary chemical molecule. Since its orderly structure is due to a maximum of stability, corresponding to a minimum of potential energy, its orderliness lacks the capacity to function as a code. The pattern of atoms forming a crystal is another instance of complex order without appreciable information content.
There is a kind of stability which often opposes the stabilizing force of a potential energy. When a liquid evaporates, this can be understood as the increase of entropy accompanying the dispersion of its particles. One takes this dispersive tendency into account by adding its powers to those of potential energy, but the correction is negligible for cases of deep drops in potential energy or for low temperatures, or for both. We can disregard it, to simplify matters, and say that chemical structures established by the stabilizing powers of chemical bonding have no appreciable information content.
In the light of the current theory of evolution, the codelike structure of DNA must be assumed to have come about by a sequence of chance variations established by natural selection. But this evolutionary aspect is irrelevant here; whatever may be the origin of a DNA configuration, it can function as a code only if its order is not due to the forces of potential energy. It must be as physically indeterminate as the sequence of words is on a printed page. As the arrangement of a printed page is extraneous to the chemistry of the printed page, so is the base sequence in a DNA molecule extraneous to the chemical forces at work in the DNA molecule. It is this physical indeterminacy of the sequence that produces the improbability of occurrence of any particular sequence and thereby enables it to have a meaning - a meaning that has a mathematically determinate information content equal to the numerical improbability of the arrangement.
DNA Acts as a Blueprint
But there remains a fundamental point to be considered. A printed page may be a mere jumble of words, and it has then no information content. So the improbability count gives the possible, rather than the actual, information content of a page. And this applies also to the information content attributed to a DNA molecule; the sequence of the bases is deemed meaningful only because we assume with Watson and Crick that this arrangement generates the structure of the offspring by endowing it with its own information content.
This brings us at last to the point that I aimed at when I undertook to analyze the information content of DNA: Can the control of morphogenesis by DNA be likened to the designing and shaping of a machine by the engineer? We have seen that physiology interprets the organism as a complex network of mechanisms, and that an organism is - like a machine - a system under dual control. Its structure is that of a boundary condition harnessing the physical-chemical substances within the organism in the service of physiological functions. Thus, in generating an organism, DNA initiates and controls the growth of a mechanism that will work as a boundary condition within a system under dual control.
And I may add that DNA itself, is such a system, since every system conveying information is under dual control, for every such system restricts and orders, in the service of conveying its information, extensive resources of particulars that would otherwise be left at random, and thereby acts as a boundary condition. In the case of DNA this boundary condition is a blueprint of the growing organism [2].
We can conclude that in each embryonic cell there is present the duplicate of a DNA molecule having a linear arrangement of its bases - an arrangement which, being independent of the chemical forces within the DNA molecules, conveys a rich amount of meaningful information. And we see that when this information is shaping the growing embryo, it produces in it boundary conditions which, themselves being independent of the physical chemical forces in which they are rooted, control the mechanism of life in the developed organism.
To elucidate this transmission is a major task of biologists today, to which 1 shall return.
1309
Some Accessory Problems Arise Here
We have seen boundary conditions introducing principles not capable of formulation in terms of physics or chemistry into inanimate artifacts and living things; we have seen them as necessary to an information content in a printed page or in DNA, and as introducing mechanical principles into machines as well as into the mechanisms of life.
Let me add now that boundary conditions of inanimate systems established by the history of the universe are found in the domains of geology, geography, and astronomy, but that these do not form systems of dual control. They resemble in this respect the test-tube type of boundaries of which I spoke above. Hence the existence of dual control in machines and living mechanisms represents a discontinuity between machines and living things on the one hand and inanimate nature on the other hand, so that both machines and living mechanisms are irreducible to the laws of physics and chemistry.
Irreducibility must not be identified with the mere fact that the joining of parts may produce features which are not observed in the separate parts. The sun is a sphere, and its parts are not spheres, nor does the law of gravitation speak of spheres; but mutual gravitational interaction causes the parts of the sun to form a sphere. Such cases of holism are common in physics and chemistry. They are often said to represent a transition to living things, but this is not the case, for they are reducible to the laws of inanimate matter, while living things are not.
But there does exist a rather different continuity between life and inanimate nature. For the beginnings of life do not sharply differ from their purely physical-chemical antecedents. One can reconcile this continuity with the irreducibility of living things by recalling the analogous case of inanimate artifacts. Take the irreducibility of machines; no animal can produce a machine, but some animals can make primitive tools, and their use of these tools may be hardly distinguishable from the mere use of the animal’s limbs. Or take a set of sounds conveying information; the set of sounds can be so obscured by noise that its presence is no longer clearly identifiable. We can say, then, that the control exercised by the boundary conditions of a system can be reduced gradually to a vanishing point. The fact that the effect of a higher principle over a system under dual control can have any value down to zero may allow us also to conceive of the continuous emergence of irreducible principles within the origin of life.
We Can Now Recognize Additional Irreducible Principles
The irreducibility of machines and printed communications teaches us, also, that the control of a system by irreducible boundary conditions does not interfere with the laws of physics and chemistry. A system under dual control relies, in fact, for the operations of its higher principle, on the working of principles of a lower level, such as the laws of physics and chemistry. Irreducible higher principles are additional to the laws of physics and chemistry. The principles of mechanical engineering and of communication of information, and the equivalent biological principles, are all additional to the laws of physics and chemistry.
But to assign the rise of such additional controlling principles to a selective process of evolution leaves serious difficulties. The production of boundary conditions in the growing fetus by transmitting to it the information contained in DNA presents a problem. Growth of a blueprint into the complex machinery that it describes seems to require a system of causes not specifiable in terms of physics and chemistry, such causes being additional both to the boundary conditions of DNA and to the morphological structure brought about by DNA.
This missing principle which builds a bodily structure on the lines of an instruction given by DNA may be exemplified by the far-reaching regenerative powers of the embryonic sea urchin, discovered by Driesch, and by Paul Weiss’s discovery that completely dispersed embryonic cells will grow, when lumped together, into a fragment of the organ from which they were isolated [3]. We see an integrative power at work here, characterized by Spemann. and by Paul Weiss as a “field” [4], which guides the growth of embryonic fragments to form the morphological features to which they embryologically belong. These guides of morphogenesis are given a formal expression in Waddington’s “epigenetic landscapes” [5]. They say graphically that the growth of the embryo is controlled by the gradient of potential shapes, much as the motion of a heavy body is controlled by the gradient of potential energy.
Remember how Driesch and his supporters fought for recognition that life transcends physics and chemistry, by arguing that the powers of regeneration in the sea urchin embryo were not explicable by a machinelike structure, and how the controversy has continued, along similar lines, between those who insisted that regulative (“equipotential” or “organismic”) integration was irreducible to any machinelike mechanism and was therefore irreducible also to the laws of inanimate nature. Now if, as I claim, machines and mechanical processes in living beings are themselves irreducible to physics and chemistry, the situation is changed. If mechanistic and organismic explanations are both equally irreducible to physics and chemistry, the recognition of organismic processes no longer bears the burden of standing alone as evidence for the irreducibility of living things. Once the “field”-like powers guiding regeneration and morphogenesis can be recognized without involving this major issue, I think the evidence for them will be found to be convincing.
There is evidence of irreducible principles, additional to those of morphological mechanisms, in the sentience that we ourselves experience and that we observe indirectly in higher animals. Most biologists set aside these matters as unprofitable considerations. But again, once it is recognized, on other grounds, that life transcends physics and chemistry, there is no reason for suspending recognition of the obvious fact that consciousness is a principle that fundamentally transcends not only physics and chemistry but also the mechanistic principles of living beings.
Biological Hierarchies Consist of a Series of Boundary Conditions
The theory of boundary conditions recognizes the higher levels of life as forming a hierarchy, each level of which relies for its workings on the principles of the levels below it, even while it itself is irreducible to these lower principles. I shall illustrate the structure of such a hierarchy by showing the way five levels make up a spoken literary composition.
The lowest level is the production of a voice; the second, the utterance of words; the third, the joining of words to make sentences; the fourth, the working of sentences into a style; the fifth,
1310
and highest, the composition of the text.
The principles of each level operate under the control of the next-higher level. The voice you produce is shaped into words by a vocabulary; a given vocabulary is shaped into sentences in accordance with a grammar; and the sentences are fitted into a style, which in turn is made to convey the ideas of the composition. Thus each level is subject to dual control: (i) control in accordance with the laws that apply to its elements in themselves, and (ii) control in accordance with the laws of the powers that control the comprehensive entity formed by these elements.
Such multiple control is made possible by the fact that the principles governing the isolated particulars of a lower level leave indeterminate conditions to be controlled by a higher principle. Voice production leaves largely open the combination of sounds into words, which is controlled by a vocabulary. Next, a vocabulary leaves largely open the combination of words to form sentences, which is controlled by grammar, and so on. Consequently, the operations of a higher level cannot be accounted for by the laws governing its particulars on the next-lower level. You cannot derive a vocabulary from phonetics; you cannot derive grammar from a vocabulary; a correct use of grammar does not account for good style; and a good style does not supply the content of a piece of prose.
Living beings comprise a whole sequence of levels forming such a hierarchy. Processes at the lowest level are caused by the forces of inanimate nature, and the higher levels control, throughout, the boundary conditions left open by the laws of inanimate nature. The lowest functions of life are those called vegetative. These vegetative functions, sustaining life at its lowest level, leave open - both in plants and in animals - the higher functions of growth and in animals also leave open the operations of muscular actions. Next, in turn, the principles governing muscular actions in animals leave open the integration of such actions to innate patterns of behavior; and, again, such patterns are open in their turn to be shaped by intelligence, while intelligence itself can be made to serve in man the still higher principles of a responsible choice.
Each level relies for its operations on all the levels below it. Each reduces the scope of the one immediately below it by imposing on it a boundary that harnesses it to the service of the next-higher level, and this control is transmitted stage by stage, down to the basic inanimate level.
The principles additional to the domain of inanimate nature are the product of an evolution the most primitive stages of which show only vegetative functions. This evolutionary progression is usually described as an increasing complexity and increasing capacity for keeping the state of the body independent of its surroundings. But if we accept, as I do, the view that living beings form a hierarchy in which each higher level represents a distinctive principle that harnesses the level below it (while being itself irreducible to its lower principles), then the evolutionary sequence gains a new and deeper significance. We can recognize then a strictly defined progression, rising from the inanimate level to ever higher additional principles of life.
This is not to say that the higher levels of life are altogether absent in earlier stages of evolution. They may be present in traces long before they become prominent. Evolution may be seen, then, as a progressive intensification of the higher principles of life. This is what we witness in the development of the embryo and of the growing child - processes akin to evolution.
But this hierarchy of principles raises once more a serious difficulty. It seems impossible to imagine that the sequence of higher principles, transcending further at each stage the laws of inanimate nature, is incipiently present in DNA and ready to be transmitted by it to the offspring. The conception of a blueprint fails to account for the transmission of faculties, like consciousness, which no mechanical device can possess. It is as if the faculty of vision were to be made intelligible to a person born blind by a chapter of sense physiology. It appears, then, that DNA evokes the ontogenesis of higher levels, rather than determining these levels. And it would follow that the emergence of the kind of hierarchy I have defined here can be only evoked, and not determined, by atomic or molecular accidents. However, this question cannot be argued here.
Understanding a Hierarchy Needs “from-at” Conceptions
I said above that the transcendence of atomism by mechanism is reflected in the fact that the presence of a mechanism is not revealed by its physical-chemical topography. We can say the same thing of all higher levels: their description in terms of any lower level does not tell us of their presence. We can generally descend to the components of a lower level by analyzing a higher level, but the opposite process involves an integration of the principles of the lower level, and this integration may be beyond our powers.
In practice this difficulty may be avoided. To take a common example, suppose that we have repeated a particular word, closely attending to the sound we are making, until these sounds have lost their meaning for us; we can recover this meaning promptly by evoking the context in which the word is commonly used. Consecutive acts of analyzing and integrating are in fact generally used for deepening our understanding of complex entities comprising two or more levels.
Yet the strictly logical difference between two consecutive levels remains. You can look at a text in a language you do not understand and see the letters that form it without being aware of their meaning, but you cannot read a text without seeing the letters that convey its meaning. This shows us two different and mutually exclusive ways of being aware of the text. When we look at words without understanding them we are focusing our attention on them, whereas, when we read the words, our attention is directed to their meaning as part of a language. We are aware then of the words only subsidiarily, as we attend to their meaning. So in the first case we are looking at the words, while in the second we are looking from them at their meaning: the reader of a text has a from-at knowledge of the words’ meaning, while he has only a from awareness of the words he is reading. Should he be able to shift his attention fully toward the words, these would lose their linguistic meaning for him.
Thus a boundary condition which harnesses the principles of a lower level in the service of a new, higher level establishes a semantic relation between the two levels. The higher comprehends the workings of the lower and thus forms the meaning of the lower. And as we ascend a hierarchy of boundaries, we reach to ever higher levels of meaning. Our understanding of the whole hierarchic edifice keeps deepening as we move upward from stage to stage.
1311
The Sequence of Boundaries Bears on Our Scientific Outlook
The recognition of a whole sequence of irreducible principles transforms the logical steps for understanding the universe of living beings. The idea, which comes to us from Galileo and Gassendi, that all manner of things must ultimately be understood in terms of matter in motion is refuted. The spectacle of physical matter forming the basic tangible ground of the universe is found to be almost empty of meaning. The universal topography of atomic particles (with their velocities and forces) which, according to Laplace, offers us a universal knowledge of all things is seen to contain hardly any knowledge that is of interest. The claims made, following the discovery of DNA, to the effect that all study of life could be reduced eventually to molecular biology, have shown once more that the Laplacean idea of universal knowledge is still the theoretical ideal of the natural sciences; current opposition to these declarations has often seemed to confirm this ideal, by defending the study of the whole organism as being only a temporary approach. But now the analysis of the hierarchy of living things shows that to reduce this hierarchy to ultimate particulars is to wipe out our very sight of it. Such analysis proves this ideal to be both false and destructive.
Each separate level of existence is of course interesting in itself and can be studied in itself. Phenomenology has taught this, by showing how to save higher, less tangible levels of experience by not trying to interpret them in terms of the more tangible things in which their existence is rooted. This method was intended to prevent the reduction of man’s mental existence to mechanical structures. The results of the method were abundant and are still flowing, but phenomenology left the ideal of exact science untouched and thus failed to secure the exclusion of its claims. Thus, phenomenological studies remained suspended over an abyss of reductionism. Moreover, the relation of the higher principles to the workings of the lowest levels in which they are rooted was lost from sight altogether.
I have mentioned how a hierarchy controlled by a series of boundary principles should be studied. When examining any higher level, we must remain subsidiarily aware of its grounds in lower levels and, turning our attention to the latter, we must continue to see them as bearing on the levels above them. Such alternation of detailing and integrating admittedly leaves open many dangers. Detailing may lead to pedantic excesses, while too-broad integrations may present us with a meandering impressionism. But the principle of stratified relations does offer at least a rational framework for an inquiry into living things and the products of human thought.
I have said that the analytic descent from higher levels to their subsidiaries is usually feasible to some degree, while the integration of items of a lower level so as to predict their possible meaning in a higher context may be beyond the range of our integrative powers. I may add now that the same things may be seen to have a joint meaning when viewed from one point, but to lack this connection when seen from another point. From an airplane we can see the traces of prehistoric sites which, over the centuries, have been unnoticed by people walking over them; indeed, once be has landed, the pilot himself may no longer see these traces.
The relation of mind to body has a similar structure. The mind-body problem arises from the disparity between the experience of a person observing an external object - for example, a cat - and a neurophysiologist observing the bodily mechanism by means of which the person sees the cat. The difference arises from the fact that the person observing the cat has a from-knowledge of the bodily responses evoked by the light in his sensory organs, and this from-knowledge integrates the joint meaning of these responses to form the sight of the cat, whereas the neurophysiologist, looking at these responses from outside, has only an at-knowledge of them, which, as such, is not integrated to form the sight of the cat. This is the same duality that exists between the airman and the pedestrian in interpreting the same traces, and the same that exists between a person who, when reading a written sentence, sees its meaning and another person who, being ignorant of the language, sees only the writing.
Awareness of mind and body confront us, therefore, with two different things. The mind harnesses neurophysiological mechanisms and is not determined by them. Owing to the existence of two kinds of awareness - the focal and the subsidiary - we can now distinguish sharply between the mind as a “from-at” experience and the subsidiaries of this experience, seen focally as a bodily mechanism. We can see then that, though rooted in the body, the mind is free in its actions - exactly as our common sense knows it to be free. The mind itself includes an ascending sequence of principles. Its appetitive and intellectual workings are transcended by principles of responsibility. Thus the growth of man to his highest levels is seen to take place along a sequence of rising principles. And we see this evolutionary hierarchy built as a sequence of boundaries, each opening the way to higher achievements by harnessing the strata below them, to which they themselves are not reducible. These boundaries control a rising series of relations which we can understand only by being aware of their constituent parts subsidiarily, as bearing on the upper level which they serve.
The recognition of certain basic impossibilities has laid the foundations of some major principles of physics and chemistry; similarly, recognition of the impossibility of understanding living things in terms of physics and chemistry, far from setting limits to our understanding of life, will guide it in the right direction. And even if the demonstration of this impossibility should prove of no great advantage in the pursuit of discovery, such a demonstration would help to draw a truer image of life and man than that given us by the present basic concepts of biology.
Summary
Mechanisms, whether man-made or morphological, are boundary conditions harnessing the laws of inanimate nature, being themselves irreducible to those laws. The pattern of organic bases in DNA which functions as a genetic code is a boundary condition irreducible to physics and chemistry. Further controlling principles of life may be represented as a hierarchy of boundary conditions extending, in the case of man, to consciousness and responsibility.
References and Notes
1. More precisely, each item consists of one out of four alternatives consisting in two positions of two different compound organic bases.
2. The blueprint carried by the DNA molecule of a particular zygote also prescribes individual features of this organism, which contribute to the sources of selective evolution, but I shall set these features aside here.
3. See P. Weiss, Proc. Nat. Acad. Sc!. U.S. 42, 819 (1956).
4. The “field” concept was first used by Spemann (1921) in describing the organizer; Paul Weiss (1923) introduced it for the study of regeneration and extended it (1926) to include ontogeny. See P. Weiss, Principles of Develop ment (Holt, New York, 1939), P. 290.
5. See, for example, C. H. Waddington, The Strategy of the Genes (Allen & Unwin, London, 1957), particularly the graphic explanation of “genetic assimilation” on page 167.
6. See, for example, M. Polanyi, Amer. Psychologist 23 (Jan. 1968) or The Tacit Dimension (Doubleday, New York, 1967).
1312
Live mechanisms and information in DNA are boundary conditions with a sequence of boundaries above them
Content
Introduction
Living Mechanisms Are Classed with Machines
DNA Information Generates Mechanisms
DNA Acts as a Blueprint
Some Accessory Problems Arise Here
We Can Now Recognize Additional Irreducible Principles
Biological Hierarchies Consist of a Series of Boundary Conditions
Understanding a Hierarchy Needs “from-at” Conceptions
The Sequence of Boundaries Bears on Our Scientific Outlook
Summary
References and Notes
Science, New Series, 160 (3834)
June 21, 1968, 1308-1312.
Introduction
If all men were exterminated, this would not affect the laws of inanimate nature. But the production of machines would stop, and not until men arose again could machines be formed once more. Some animals can produce tools, but only men can construct machines; machines are human artifacts, made of inanimate material.
The Oxford Dictionary describes a machine as “an apparatus for applying mechanical power, consisting of a number of interrelated parts, each having a definite function.” It might be, for example, a machine for sewing or printing. Let us assume that the power driving the machine is built in, and disregard the fact that it has to be renewed from time to time. We can say, then, that the manufacture of a machine consists in cutting suitably shaped parts and fitting them together so that their joint mechanical action should serve a possible human purpose.
The structure of machines and the working of their structure are thus shaped by man, even while their material and the forces that operate them obey the laws of inanimate nature. In constructing a machine and supplying it with power, we harness the laws of nature at work in its material and in its driving force and make them serve our purpose.
This harness is not unbreakable; the structure of the machine, and thus its working, can break down. But this will not affect the forces of inanimate nature on which the operation of the machine relied; it merely releases them from the restriction the machine imposed on them before it broke down.
So the machine as a whole works under the control of two distinct principles. The higher one is the principle of the machine’s design, and this harnesses the lower one, which consists in the physical-chemical processes on which the machine relies. We commonly form such a two-leveled structure in conducting an experiment; but there is a difference between constructing a machine and rigging up an experiment. The experimenter imposes restrictions on nature in order to observe its behavior under these restrictions, while the constructor of a machine restricts nature in order to harness its workings. But we may borrow a term from physics and describe both these useful restrictions of nature as the imposing of boundary conditions on the laws of physics and chemistry.
Let me enlarge on this. I have exemplified two types of boundaries. In the machine our principal interest lay in the effects of the boundary conditions, while in an experimental setting we are interested in the natural processes controlled by the boundaries. There are many common examples of both types of boundaries. When a saucepan bounds a soup that we are cooking, we are interested in the soup; and, likewise, when we observe a reaction in a test tube, we are studying the reaction, not the test tube. The reverse is true for a game of chess. The strategy of the player imposes boundaries on the several moves, which follow the laws of chess, but our interest lies in the boundaries - that is, in the strategy, not in the several moves as exemplifications of the laws. And similarly, when a sculptor shapes a stone or a painter composes a painting, our interest lies in the boundaries imposed on a material, and not in the material itself.
We can distinguish these two types of boundaries by saying that the first represents a test-tube type of boundary whereas the second is of the machine type. By shifting our attention, we may sometimes change a boundary from one type to another.
All communications form a machine type of boundary, and these boundaries form a whole hierarchy of consecutive levels of action. A vocabulary sets boundary conditions on the utterance of the spoken voice; a grammar harnesses words to form sentences, and the sentences are shaped into a text which conveys a communication. At all these stages we are interested in the boundaries imposed by a comprehensive restrictive power, rather than in the principles harnessed by them.
Living Mechanisms Are Classed with Machines
From machines we pass to living beings, by remembering that animals move about mechanically and that they have internal organs which perform functions as parts of a machine do - functions which sustain the life of the organism, much as the proper functioning of parts of a machine keeps the machine going. For centuries past, the workings of life have been likened to the working of machines and physiology has been seeking to interpret the organism as a complex network of mechanisms. Organs are, accordingly, defined by their life-preserving functions.
Any coherent part of the organism is indeed puzzling to physiology - and also meaningless to pathology - until the way it benefits the organism is discovered. And I may add that any description of such a system in terms of its physical-chemical topography is meaningless, except for the fact that the description covertly may recall the system’s physiological interpretation - much as the topography of a machine is meaningless until we guess how the device works, and for what purpose.
In this light the organism is shown to be, like a machine, a system which works according to two different principles: its structure serves as a boundary condition harnessing the physical-chemical processes by which its organs perform their functions. Thus, this system may be called a system under dual control. Morphogenesis, the process by which the structure of living beings develops, can then be likened to the shaping of a machine which will act as a boundary for the laws of inanimate nature. For just as these laws serve the machine, so they serve also the developed organism.
* The author is a former Fellow of Merton College, Oxford, and Emeritus Professor of social studies at the University of Manchester, where he had previously held the Chair of Physical Chemistry. His present address is 22 Upland Park Road, Oxford, England. This article is an expanded version of a paper presented 20 December 1967 at the New York meeting of the AAAS. The first half of the article was anticipated in a paper published in the August 1967 issue of Chemical and Engineering News.
1308
A boundary condition is always extraneous to the process which it delimits. In Galileo’s experiments on balls rolling down a slope, the angle of the slope was not derived from the laws of mechanics, but was chosen by Galileo. And as this choice of slopes was extraneous to the laws of mechanics, so is the shape and manufacture of test tubes extraneous to the laws of chemistry.
The same thing holds for machine-like boundaries; their structure cannot be defined in terms of the laws which they harness. Nor can a vocabulary determine the content of a text, and so on. Therefore, if the structure of living things is a set of boundary conditions, this structure is extraneous to the laws of physics and chemistry which the organism is harnessing. Thus the morphology of living things transcends the laws of physics and chemistry.
DNA Information Generates Mechanisms
But the analogy between machine components and live functioning organs is weakened by the fact that the organs are not shaped artificially as the parts of a machine are. It is an advantage, therefore, to find that the morphogenetic process is explained in principle by the transmission of information stored in DNA, interpreted in this sense by Watson and Crick.
A DNA molecule is said to represent a code - that is, a linear sequence of items, the arrangement of which is the information conveyed by the code. In the case of DNA, each item of the series consists of one out of four alternative organic bases [1]. Such a code will convey the maximum amount of information if the four organic bases have equal probability of forming any particular item of the series. Any difference in the binding of the four alternative bases, whether at the same point of the series or between two points of the series, will cause the information conveyed by the series to fall below the ideal maximum. The information content of DNA is in fact known to be reduced to some extent by redundancy, but I accept here the assumption of Watson and Crick that this redundancy does not prevent DNA from effectively functioning as a code. I accordingly disregard, for the sake of brevity, the redundancy in the DNA code and talk of it as if it were functioning optimally, with all of its alternative basic bindings having the same probability of occurrence.
Let us be clear what would happen in the opposite case. Suppose that the actual structure of a DNA molecule were due to the fact that the bindings of its bases were much stronger than the bindings would be for any other distribution of bases, then such a DNA molecule would have no information content. Its codelike character would be effaced by an overwhelming redundancy.
We may note that such is actually the case for an ordinary chemical molecule. Since its orderly structure is due to a maximum of stability, corresponding to a minimum of potential energy, its orderliness lacks the capacity to function as a code. The pattern of atoms forming a crystal is another instance of complex order without appreciable information content.
There is a kind of stability which often opposes the stabilizing force of a potential energy. When a liquid evaporates, this can be understood as the increase of entropy accompanying the dispersion of its particles. One takes this dispersive tendency into account by adding its powers to those of potential energy, but the correction is negligible for cases of deep drops in potential energy or for low temperatures, or for both. We can disregard it, to simplify matters, and say that chemical structures established by the stabilizing powers of chemical bonding have no appreciable information content.
In the light of the current theory of evolution, the codelike structure of DNA must be assumed to have come about by a sequence of chance variations established by natural selection. But this evolutionary aspect is irrelevant here; whatever may be the origin of a DNA configuration, it can function as a code only if its order is not due to the forces of potential energy. It must be as physically indeterminate as the sequence of words is on a printed page. As the arrangement of a printed page is extraneous to the chemistry of the printed page, so is the base sequence in a DNA molecule extraneous to the chemical forces at work in the DNA molecule. It is this physical indeterminacy of the sequence that produces the improbability of occurrence of any particular sequence and thereby enables it to have a meaning - a meaning that has a mathematically determinate information content equal to the numerical improbability of the arrangement.
DNA Acts as a Blueprint
But there remains a fundamental point to be considered. A printed page may be a mere jumble of words, and it has then no information content. So the improbability count gives the possible, rather than the actual, information content of a page. And this applies also to the information content attributed to a DNA molecule; the sequence of the bases is deemed meaningful only because we assume with Watson and Crick that this arrangement generates the structure of the offspring by endowing it with its own information content.
This brings us at last to the point that I aimed at when I undertook to analyze the information content of DNA: Can the control of morphogenesis by DNA be likened to the designing and shaping of a machine by the engineer? We have seen that physiology interprets the organism as a complex network of mechanisms, and that an organism is - like a machine - a system under dual control. Its structure is that of a boundary condition harnessing the physical-chemical substances within the organism in the service of physiological functions. Thus, in generating an organism, DNA initiates and controls the growth of a mechanism that will work as a boundary condition within a system under dual control.
And I may add that DNA itself, is such a system, since every system conveying information is under dual control, for every such system restricts and orders, in the service of conveying its information, extensive resources of particulars that would otherwise be left at random, and thereby acts as a boundary condition. In the case of DNA this boundary condition is a blueprint of the growing organism [2].
We can conclude that in each embryonic cell there is present the duplicate of a DNA molecule having a linear arrangement of its bases - an arrangement which, being independent of the chemical forces within the DNA molecules, conveys a rich amount of meaningful information. And we see that when this information is shaping the growing embryo, it produces in it boundary conditions which, themselves being independent of the physical chemical forces in which they are rooted, control the mechanism of life in the developed organism.
To elucidate this transmission is a major task of biologists today, to which 1 shall return.
1309
Some Accessory Problems Arise Here
We have seen boundary conditions introducing principles not capable of formulation in terms of physics or chemistry into inanimate artifacts and living things; we have seen them as necessary to an information content in a printed page or in DNA, and as introducing mechanical principles into machines as well as into the mechanisms of life.
Let me add now that boundary conditions of inanimate systems established by the history of the universe are found in the domains of geology, geography, and astronomy, but that these do not form systems of dual control. They resemble in this respect the test-tube type of boundaries of which I spoke above. Hence the existence of dual control in machines and living mechanisms represents a discontinuity between machines and living things on the one hand and inanimate nature on the other hand, so that both machines and living mechanisms are irreducible to the laws of physics and chemistry.
Irreducibility must not be identified with the mere fact that the joining of parts may produce features which are not observed in the separate parts. The sun is a sphere, and its parts are not spheres, nor does the law of gravitation speak of spheres; but mutual gravitational interaction causes the parts of the sun to form a sphere. Such cases of holism are common in physics and chemistry. They are often said to represent a transition to living things, but this is not the case, for they are reducible to the laws of inanimate matter, while living things are not.
But there does exist a rather different continuity between life and inanimate nature. For the beginnings of life do not sharply differ from their purely physical-chemical antecedents. One can reconcile this continuity with the irreducibility of living things by recalling the analogous case of inanimate artifacts. Take the irreducibility of machines; no animal can produce a machine, but some animals can make primitive tools, and their use of these tools may be hardly distinguishable from the mere use of the animal’s limbs. Or take a set of sounds conveying information; the set of sounds can be so obscured by noise that its presence is no longer clearly identifiable. We can say, then, that the control exercised by the boundary conditions of a system can be reduced gradually to a vanishing point. The fact that the effect of a higher principle over a system under dual control can have any value down to zero may allow us also to conceive of the continuous emergence of irreducible principles within the origin of life.
We Can Now Recognize Additional Irreducible Principles
The irreducibility of machines and printed communications teaches us, also, that the control of a system by irreducible boundary conditions does not interfere with the laws of physics and chemistry. A system under dual control relies, in fact, for the operations of its higher principle, on the working of principles of a lower level, such as the laws of physics and chemistry. Irreducible higher principles are additional to the laws of physics and chemistry. The principles of mechanical engineering and of communication of information, and the equivalent biological principles, are all additional to the laws of physics and chemistry.
But to assign the rise of such additional controlling principles to a selective process of evolution leaves serious difficulties. The production of boundary conditions in the growing fetus by transmitting to it the information contained in DNA presents a problem. Growth of a blueprint into the complex machinery that it describes seems to require a system of causes not specifiable in terms of physics and chemistry, such causes being additional both to the boundary conditions of DNA and to the morphological structure brought about by DNA.
This missing principle which builds a bodily structure on the lines of an instruction given by DNA may be exemplified by the far-reaching regenerative powers of the embryonic sea urchin, discovered by Driesch, and by Paul Weiss’s discovery that completely dispersed embryonic cells will grow, when lumped together, into a fragment of the organ from which they were isolated [3]. We see an integrative power at work here, characterized by Spemann. and by Paul Weiss as a “field” [4], which guides the growth of embryonic fragments to form the morphological features to which they embryologically belong. These guides of morphogenesis are given a formal expression in Waddington’s “epigenetic landscapes” [5]. They say graphically that the growth of the embryo is controlled by the gradient of potential shapes, much as the motion of a heavy body is controlled by the gradient of potential energy.
Remember how Driesch and his supporters fought for recognition that life transcends physics and chemistry, by arguing that the powers of regeneration in the sea urchin embryo were not explicable by a machinelike structure, and how the controversy has continued, along similar lines, between those who insisted that regulative (“equipotential” or “organismic”) integration was irreducible to any machinelike mechanism and was therefore irreducible also to the laws of inanimate nature. Now if, as I claim, machines and mechanical processes in living beings are themselves irreducible to physics and chemistry, the situation is changed. If mechanistic and organismic explanations are both equally irreducible to physics and chemistry, the recognition of organismic processes no longer bears the burden of standing alone as evidence for the irreducibility of living things. Once the “field”-like powers guiding regeneration and morphogenesis can be recognized without involving this major issue, I think the evidence for them will be found to be convincing.
There is evidence of irreducible principles, additional to those of morphological mechanisms, in the sentience that we ourselves experience and that we observe indirectly in higher animals. Most biologists set aside these matters as unprofitable considerations. But again, once it is recognized, on other grounds, that life transcends physics and chemistry, there is no reason for suspending recognition of the obvious fact that consciousness is a principle that fundamentally transcends not only physics and chemistry but also the mechanistic principles of living beings.
Biological Hierarchies Consist of a Series of Boundary Conditions
The theory of boundary conditions recognizes the higher levels of life as forming a hierarchy, each level of which relies for its workings on the principles of the levels below it, even while it itself is irreducible to these lower principles. I shall illustrate the structure of such a hierarchy by showing the way five levels make up a spoken literary composition.
The lowest level is the production of a voice; the second, the utterance of words; the third, the joining of words to make sentences; the fourth, the working of sentences into a style; the fifth,
1310
and highest, the composition of the text.
The principles of each level operate under the control of the next-higher level. The voice you produce is shaped into words by a vocabulary; a given vocabulary is shaped into sentences in accordance with a grammar; and the sentences are fitted into a style, which in turn is made to convey the ideas of the composition. Thus each level is subject to dual control: (i) control in accordance with the laws that apply to its elements in themselves, and (ii) control in accordance with the laws of the powers that control the comprehensive entity formed by these elements.
Such multiple control is made possible by the fact that the principles governing the isolated particulars of a lower level leave indeterminate conditions to be controlled by a higher principle. Voice production leaves largely open the combination of sounds into words, which is controlled by a vocabulary. Next, a vocabulary leaves largely open the combination of words to form sentences, which is controlled by grammar, and so on. Consequently, the operations of a higher level cannot be accounted for by the laws governing its particulars on the next-lower level. You cannot derive a vocabulary from phonetics; you cannot derive grammar from a vocabulary; a correct use of grammar does not account for good style; and a good style does not supply the content of a piece of prose.
Living beings comprise a whole sequence of levels forming such a hierarchy. Processes at the lowest level are caused by the forces of inanimate nature, and the higher levels control, throughout, the boundary conditions left open by the laws of inanimate nature. The lowest functions of life are those called vegetative. These vegetative functions, sustaining life at its lowest level, leave open - both in plants and in animals - the higher functions of growth and in animals also leave open the operations of muscular actions. Next, in turn, the principles governing muscular actions in animals leave open the integration of such actions to innate patterns of behavior; and, again, such patterns are open in their turn to be shaped by intelligence, while intelligence itself can be made to serve in man the still higher principles of a responsible choice.
Each level relies for its operations on all the levels below it. Each reduces the scope of the one immediately below it by imposing on it a boundary that harnesses it to the service of the next-higher level, and this control is transmitted stage by stage, down to the basic inanimate level.
The principles additional to the domain of inanimate nature are the product of an evolution the most primitive stages of which show only vegetative functions. This evolutionary progression is usually described as an increasing complexity and increasing capacity for keeping the state of the body independent of its surroundings. But if we accept, as I do, the view that living beings form a hierarchy in which each higher level represents a distinctive principle that harnesses the level below it (while being itself irreducible to its lower principles), then the evolutionary sequence gains a new and deeper significance. We can recognize then a strictly defined progression, rising from the inanimate level to ever higher additional principles of life.
This is not to say that the higher levels of life are altogether absent in earlier stages of evolution. They may be present in traces long before they become prominent. Evolution may be seen, then, as a progressive intensification of the higher principles of life. This is what we witness in the development of the embryo and of the growing child - processes akin to evolution.
But this hierarchy of principles raises once more a serious difficulty. It seems impossible to imagine that the sequence of higher principles, transcending further at each stage the laws of inanimate nature, is incipiently present in DNA and ready to be transmitted by it to the offspring. The conception of a blueprint fails to account for the transmission of faculties, like consciousness, which no mechanical device can possess. It is as if the faculty of vision were to be made intelligible to a person born blind by a chapter of sense physiology. It appears, then, that DNA evokes the ontogenesis of higher levels, rather than determining these levels. And it would follow that the emergence of the kind of hierarchy I have defined here can be only evoked, and not determined, by atomic or molecular accidents. However, this question cannot be argued here.
Understanding a Hierarchy Needs “from-at” Conceptions
I said above that the transcendence of atomism by mechanism is reflected in the fact that the presence of a mechanism is not revealed by its physical-chemical topography. We can say the same thing of all higher levels: their description in terms of any lower level does not tell us of their presence. We can generally descend to the components of a lower level by analyzing a higher level, but the opposite process involves an integration of the principles of the lower level, and this integration may be beyond our powers.
In practice this difficulty may be avoided. To take a common example, suppose that we have repeated a particular word, closely attending to the sound we are making, until these sounds have lost their meaning for us; we can recover this meaning promptly by evoking the context in which the word is commonly used. Consecutive acts of analyzing and integrating are in fact generally used for deepening our understanding of complex entities comprising two or more levels.
Yet the strictly logical difference between two consecutive levels remains. You can look at a text in a language you do not understand and see the letters that form it without being aware of their meaning, but you cannot read a text without seeing the letters that convey its meaning. This shows us two different and mutually exclusive ways of being aware of the text. When we look at words without understanding them we are focusing our attention on them, whereas, when we read the words, our attention is directed to their meaning as part of a language. We are aware then of the words only subsidiarily, as we attend to their meaning. So in the first case we are looking at the words, while in the second we are looking from them at their meaning: the reader of a text has a from-at knowledge of the words’ meaning, while he has only a from awareness of the words he is reading. Should he be able to shift his attention fully toward the words, these would lose their linguistic meaning for him.
Thus a boundary condition which harnesses the principles of a lower level in the service of a new, higher level establishes a semantic relation between the two levels. The higher comprehends the workings of the lower and thus forms the meaning of the lower. And as we ascend a hierarchy of boundaries, we reach to ever higher levels of meaning. Our understanding of the whole hierarchic edifice keeps deepening as we move upward from stage to stage.
1311
The Sequence of Boundaries Bears on Our Scientific Outlook
The recognition of a whole sequence of irreducible principles transforms the logical steps for understanding the universe of living beings. The idea, which comes to us from Galileo and Gassendi, that all manner of things must ultimately be understood in terms of matter in motion is refuted. The spectacle of physical matter forming the basic tangible ground of the universe is found to be almost empty of meaning. The universal topography of atomic particles (with their velocities and forces) which, according to Laplace, offers us a universal knowledge of all things is seen to contain hardly any knowledge that is of interest. The claims made, following the discovery of DNA, to the effect that all study of life could be reduced eventually to molecular biology, have shown once more that the Laplacean idea of universal knowledge is still the theoretical ideal of the natural sciences; current opposition to these declarations has often seemed to confirm this ideal, by defending the study of the whole organism as being only a temporary approach. But now the analysis of the hierarchy of living things shows that to reduce this hierarchy to ultimate particulars is to wipe out our very sight of it. Such analysis proves this ideal to be both false and destructive.
Each separate level of existence is of course interesting in itself and can be studied in itself. Phenomenology has taught this, by showing how to save higher, less tangible levels of experience by not trying to interpret them in terms of the more tangible things in which their existence is rooted. This method was intended to prevent the reduction of man’s mental existence to mechanical structures. The results of the method were abundant and are still flowing, but phenomenology left the ideal of exact science untouched and thus failed to secure the exclusion of its claims. Thus, phenomenological studies remained suspended over an abyss of reductionism. Moreover, the relation of the higher principles to the workings of the lowest levels in which they are rooted was lost from sight altogether.
I have mentioned how a hierarchy controlled by a series of boundary principles should be studied. When examining any higher level, we must remain subsidiarily aware of its grounds in lower levels and, turning our attention to the latter, we must continue to see them as bearing on the levels above them. Such alternation of detailing and integrating admittedly leaves open many dangers. Detailing may lead to pedantic excesses, while too-broad integrations may present us with a meandering impressionism. But the principle of stratified relations does offer at least a rational framework for an inquiry into living things and the products of human thought.
I have said that the analytic descent from higher levels to their subsidiaries is usually feasible to some degree, while the integration of items of a lower level so as to predict their possible meaning in a higher context may be beyond the range of our integrative powers. I may add now that the same things may be seen to have a joint meaning when viewed from one point, but to lack this connection when seen from another point. From an airplane we can see the traces of prehistoric sites which, over the centuries, have been unnoticed by people walking over them; indeed, once be has landed, the pilot himself may no longer see these traces.
The relation of mind to body has a similar structure. The mind-body problem arises from the disparity between the experience of a person observing an external object - for example, a cat - and a neurophysiologist observing the bodily mechanism by means of which the person sees the cat. The difference arises from the fact that the person observing the cat has a from-knowledge of the bodily responses evoked by the light in his sensory organs, and this from-knowledge integrates the joint meaning of these responses to form the sight of the cat, whereas the neurophysiologist, looking at these responses from outside, has only an at-knowledge of them, which, as such, is not integrated to form the sight of the cat. This is the same duality that exists between the airman and the pedestrian in interpreting the same traces, and the same that exists between a person who, when reading a written sentence, sees its meaning and another person who, being ignorant of the language, sees only the writing.
Awareness of mind and body confront us, therefore, with two different things. The mind harnesses neurophysiological mechanisms and is not determined by them. Owing to the existence of two kinds of awareness - the focal and the subsidiary - we can now distinguish sharply between the mind as a “from-at” experience and the subsidiaries of this experience, seen focally as a bodily mechanism. We can see then that, though rooted in the body, the mind is free in its actions - exactly as our common sense knows it to be free. The mind itself includes an ascending sequence of principles. Its appetitive and intellectual workings are transcended by principles of responsibility. Thus the growth of man to his highest levels is seen to take place along a sequence of rising principles. And we see this evolutionary hierarchy built as a sequence of boundaries, each opening the way to higher achievements by harnessing the strata below them, to which they themselves are not reducible. These boundaries control a rising series of relations which we can understand only by being aware of their constituent parts subsidiarily, as bearing on the upper level which they serve.
The recognition of certain basic impossibilities has laid the foundations of some major principles of physics and chemistry; similarly, recognition of the impossibility of understanding living things in terms of physics and chemistry, far from setting limits to our understanding of life, will guide it in the right direction. And even if the demonstration of this impossibility should prove of no great advantage in the pursuit of discovery, such a demonstration would help to draw a truer image of life and man than that given us by the present basic concepts of biology.
Summary
Mechanisms, whether man-made or morphological, are boundary conditions harnessing the laws of inanimate nature, being themselves irreducible to those laws. The pattern of organic bases in DNA which functions as a genetic code is a boundary condition irreducible to physics and chemistry. Further controlling principles of life may be represented as a hierarchy of boundary conditions extending, in the case of man, to consciousness and responsibility.
References and Notes
1. More precisely, each item consists of one out of four alternatives consisting in two positions of two different compound organic bases.
2. The blueprint carried by the DNA molecule of a particular zygote also prescribes individual features of this organism, which contribute to the sources of selective evolution, but I shall set these features aside here.
3. See P. Weiss, Proc. Nat. Acad. Sc!. U.S. 42, 819 (1956).
4. The “field” concept was first used by Spemann (1921) in describing the organizer; Paul Weiss (1923) introduced it for the study of regeneration and extended it (1926) to include ontogeny. See P. Weiss, Principles of Develop ment (Holt, New York, 1939), P. 290.
5. See, for example, C. H. Waddington, The Strategy of the Genes (Allen & Unwin, London, 1957), particularly the graphic explanation of “genetic assimilation” on page 167.
6. See, for example, M. Polanyi, Amer. Psychologist 23 (Jan. 1968) or The Tacit Dimension (Doubleday, New York, 1967).
1312
вторник, 15 марта 2011 г.
Существование коллективного разума на бирже доказали научно
Трейдеры, синхронизирующие свои решения с другими участниками торгов, к концу дня чаще остаются в выигрыше по сравнению с теми, кто действует без оглядки на остальных. Такой вывод о существовании "коллективного разума" сделала группа ученых, наблюдавшая за поведением участников торгов. Работа исследователей опубликована в журнале Proceedings of the National Academy of Sciences, а коротко о ней пишет портал Science News.
Авторы работы анализировали торговые операции, проводимые 66 трейдерами на бирже в течение полутора лет. В общей сложности, ученые рассмотрели более миллиона сделок. Многие трейдеры постоянно общаются между собой - например, путем передачи сообщений при помощи различных сервисов по обмену мгновенными сообщениями.
Ученые определили, что основной пик обмена такими сообщениями наблюдается в период от нескольких секунд до нескольких минут после появления какой-либо информации, способной повлиять на изменение цен акций. Более того, оказалось, что трейдеры, взаимодействующие с коллегами и принимающие решения, которые выбирает большинство, получают прибыль, в среднем, в 55 процентах сделок (в некоторых случаях этот показатель возрастал до 60 процентов).
Коллеги авторов хорошо восприняли их работу, отметив ее контринтуитивный результат - обычно считается, что стратегия успеха на бирже включает необычные решения - не такие, как выбирает большинство участников торгов. Новая работы показывает, что плыть по течению во многих случаях оказывается выгодно.
Недавно другой коллектив ученых вывел еще одну закономерность, связанную с торгами на бирже - исследователи смогли установить корреляцию между эмоциональной окраской постов в микроблоге Twitter и колебаниями индекса Dow Jones.
Ссылки по теме
- Synchronicity, instant messaging, and performance among financial traders - Proceedings of the National Academy of Sciences, 14.03.2011
- Jumping on the bandwagon brings rewards - Science News, 14.03.2011
- Швейцарские ученые попробуют себя в роли экономических оракулов – Lenta.ru, 05.11.2009
- Эконофизики раскрыли метод предсказания падения финансовых индексов – Lenta.ru, 10.09.2009
- Ученый составил карту распространения финансового кризиса – Lenta.ru, 14.01.2009
- Физики научились предсказывать цены на акции по птичьему полету – Lenta.ru, 10.08.2007
Трейдеры, синхронизирующие свои решения с другими участниками торгов, к концу дня чаще остаются в выигрыше по сравнению с теми, кто действует без оглядки на остальных. Такой вывод о существовании "коллективного разума" сделала группа ученых, наблюдавшая за поведением участников торгов. Работа исследователей опубликована в журнале Proceedings of the National Academy of Sciences, а коротко о ней пишет портал Science News.
Авторы работы анализировали торговые операции, проводимые 66 трейдерами на бирже в течение полутора лет. В общей сложности, ученые рассмотрели более миллиона сделок. Многие трейдеры постоянно общаются между собой - например, путем передачи сообщений при помощи различных сервисов по обмену мгновенными сообщениями.
Ученые определили, что основной пик обмена такими сообщениями наблюдается в период от нескольких секунд до нескольких минут после появления какой-либо информации, способной повлиять на изменение цен акций. Более того, оказалось, что трейдеры, взаимодействующие с коллегами и принимающие решения, которые выбирает большинство, получают прибыль, в среднем, в 55 процентах сделок (в некоторых случаях этот показатель возрастал до 60 процентов).
Коллеги авторов хорошо восприняли их работу, отметив ее контринтуитивный результат - обычно считается, что стратегия успеха на бирже включает необычные решения - не такие, как выбирает большинство участников торгов. Новая работы показывает, что плыть по течению во многих случаях оказывается выгодно.
Недавно другой коллектив ученых вывел еще одну закономерность, связанную с торгами на бирже - исследователи смогли установить корреляцию между эмоциональной окраской постов в микроблоге Twitter и колебаниями индекса Dow Jones.
Ссылки по теме
- Synchronicity, instant messaging, and performance among financial traders - Proceedings of the National Academy of Sciences, 14.03.2011
- Jumping on the bandwagon brings rewards - Science News, 14.03.2011
- Швейцарские ученые попробуют себя в роли экономических оракулов – Lenta.ru, 05.11.2009
- Эконофизики раскрыли метод предсказания падения финансовых индексов – Lenta.ru, 10.09.2009
- Ученый составил карту распространения финансового кризиса – Lenta.ru, 14.01.2009
- Физики научились предсказывать цены на акции по птичьему полету – Lenta.ru, 10.08.2007
суббота, 12 марта 2011 г.
Американские ученые предложили победить вымирание истреблением
Американские ученые предложили необычный способ борьбы с вымиранием. Статья исследователей появилась в журнале Nature Communications. Ее препринт доступен на сайте arXiv.org.
В рамках работы ученые изучали так называемое каскадное вымирание: исчезновение (или резкое сокращение популяции) некоторого вида приводит к гибели тех видов, которые им питались, и так далее по пищевой цепочке. В качестве моделей исследователи использовали несколько реально существующих пищевых систем (включая, например, пищевую цепочку Чесапикского залива, куда входит 33 вида), а также некоторое количество сгенерированных на компьютере.
Исследователи намеревались выяснить воздействие дополнительных возмущений пищевой системы в случае, когда вымирание уже началось. В результате ученым удалось установить, что подобные возмущения не только не усугубляли ситуацию, но в некоторых случаях позволяли остановить каскадное вымирание. По словам исследователей, подобный подход был впервые опробован для сложных экосистем, которые они изучали в работе.
В качестве одного из вариантов подобных возмущений исследователи предложили намеренное снижение (вплоть до полного истребления ) популяции некоторых видов в пищевой цепи, стоящих "на пути" каскадного вымирания. Ученые полагают, что, несмотря на всю неоднозначность, новый метод поможет в контроле популяции реальных экосистем.
Американские ученые предложили необычный способ борьбы с вымиранием. Статья исследователей появилась в журнале Nature Communications. Ее препринт доступен на сайте arXiv.org.
В рамках работы ученые изучали так называемое каскадное вымирание: исчезновение (или резкое сокращение популяции) некоторого вида приводит к гибели тех видов, которые им питались, и так далее по пищевой цепочке. В качестве моделей исследователи использовали несколько реально существующих пищевых систем (включая, например, пищевую цепочку Чесапикского залива, куда входит 33 вида), а также некоторое количество сгенерированных на компьютере.
Исследователи намеревались выяснить воздействие дополнительных возмущений пищевой системы в случае, когда вымирание уже началось. В результате ученым удалось установить, что подобные возмущения не только не усугубляли ситуацию, но в некоторых случаях позволяли остановить каскадное вымирание. По словам исследователей, подобный подход был впервые опробован для сложных экосистем, которые они изучали в работе.
В качестве одного из вариантов подобных возмущений исследователи предложили намеренное снижение (вплоть до полного истребления ) популяции некоторых видов в пищевой цепи, стоящих "на пути" каскадного вымирания. Ученые полагают, что, несмотря на всю неоднозначность, новый метод поможет в контроле популяции реальных экосистем.
четверг, 27 января 2011 г.
суббота, 22 января 2011 г.
The Truth Wears Off
Is there something wrong with the scientific method?by Jonah Lehrer
December 13, 2010
Read more http://www.newyorker.com/reporting/2010/12/13/101213fa_fact_lehrer#ixzz1Bl8DSHE5
The Truth Wears OffIs there something wrong with the scientific method?by Jonah Lehrer
December 13, 2010 Many results that are rigorously proved and accepted start shrinking in later studies.
Share Print E-Mail Single Page Keywords
Scientific Experiments; Decline Effect; Replicability; Scientists; Statistics; Jonathan Schooler; Scientific Theories On September 18, 2007, a few dozen neuroscientists, psychiatrists, and drug-company executives gathered in a hotel conference room in Brussels to hear some startling news. It had to do with a class of drugs known as atypical or second-generation antipsychotics, which came on the market in the early nineties. The drugs, sold under brand names such as Abilify, Seroquel, and Zyprexa, had been tested on schizophrenics in several large clinical trials, all of which had demonstrated a dramatic decrease in the subjects’ psychiatric symptoms. As a result, second-generation antipsychotics had become one of the fastest-growing and most profitable pharmaceutical classes. By 2001, Eli Lilly’s Zyprexa was generating more revenue than Prozac. It remains the company’s top-selling drug.
But the data presented at the Brussels meeting made it clear that something strange was happening: the therapeutic power of the drugs appeared to be steadily waning. A recent study showed an effect that was less than half of that documented in the first trials, in the early nineteen-nineties. Many researchers began to argue that the expensive pharmaceuticals weren’t any better than first-generation antipsychotics, which have been in use since the fifties. “In fact, sometimes they now look even worse,” John Davis, a professor of psychiatry at the University of Illinois at Chicago, told me.
Before the effectiveness of a drug can be confirmed, it must be tested and tested again. Different scientists in different labs need to repeat the protocols and publish their results. The test of replicability, as it’s known, is the foundation of modern research. Replicability is how the community enforces itself. It’s a safeguard for the creep of subjectivity. Most of the time, scientists know what results they want, and that can influence the results they get. The premise of replicability is that the scientific community can correct for these flaws.
But now all sorts of well-established, multiply confirmed findings have started to look increasingly uncertain. It’s as if our facts were losing their truth: claims that have been enshrined in textbooks are suddenly unprovable. This phenomenon doesn’t yet have an official name, but it’s occurring across a wide range of fields, from psychology to ecology. In the field of medicine, the phenomenon seems extremely widespread, affecting not only antipsychotics but also therapies ranging from cardiac stents to Vitamin E and antidepressants: Davis has a forthcoming analysis demonstrating that the efficacy of antidepressants has gone down as much as threefold in recent decades.
For many scientists, the effect is especially troubling because of what it exposes about the scientific process. If replication is what separates the rigor of science from the squishiness of pseudoscience, where do we put all these rigorously validated findings that can no longer be proved? Which results should we believe? Francis Bacon, the early-modern philosopher and pioneer of the scientific method, once declared that experiments were essential, because they allowed us to “put nature to the question.” But it appears that nature often gives us different answers.
from the issuecartoon banke-mail thisJonathan Schooler was a young graduate student at the University of Washington in the nineteen-eighties when he discovered a surprising new fact about language and memory. At the time, it was widely believed that the act of describing our memories improved them. But, in a series of clever experiments, Schooler demonstrated that subjects shown a face and asked to describe it were much less likely to recognize the face when shown it later than those who had simply looked at it. Schooler called the phenomenon “verbal overshadowing.”
The study turned him into an academic star. Since its initial publication, in 1990, it has been cited more than four hundred times. Before long, Schooler had extended the model to a variety of other tasks, such as remembering the taste of a wine, identifying the best strawberry jam, and solving difficult creative puzzles. In each instance, asking people to put their perceptions into words led to dramatic decreases in performance.
But while Schooler was publishing these results in highly reputable journals, a secret worry gnawed at him: it was proving difficult to replicate his earlier findings. “I’d often still see an effect, but the effect just wouldn’t be as strong,” he told me. “It was as if verbal overshadowing, my big new idea, was getting weaker.” At first, he assumed that he’d made an error in experimental design or a statistical miscalculation. But he couldn’t find anything wrong with his research. He then concluded that his initial batch of research subjects must have been unusually susceptible to verbal overshadowing. (John Davis, similarly, has speculated that part of the drop-off in the effectiveness of antipsychotics can be attributed to using subjects who suffer from milder forms of psychosis which are less likely to show dramatic improvement.) “It wasn’t a very satisfying explanation,” Schooler says. “One of my mentors told me that my real mistake was trying to replicate my work. He told me doing that was just setting myself up for disappointment.”
Schooler tried to put the problem out of his mind; his colleagues assured him that such things happened all the time. Over the next few years, he found new research questions, got married and had kids. But his replication problem kept on getting worse. His first attempt at replicating the 1990 study, in 1995, resulted in an effect that was thirty per cent smaller. The next year, the size of the effect shrank another thirty per cent. When other labs repeated Schooler’s experiments, they got a similar spread of data, with a distinct downward trend. “This was profoundly frustrating,” he says. “It was as if nature gave me this great result and then tried to take it back.” In private, Schooler began referring to the problem as “cosmic habituation,” by analogy to the decrease in response that occurs when individuals habituate to particular stimuli. “Habituation is why you don’t notice the stuff that’s always there,” Schooler says. “It’s an inevitable process of adjustment, a ratcheting down of excitement. I started joking that it was like the cosmos was habituating to my ideas. I took it very personally.”
Schooler is now a tenured professor at the University of California at Santa Barbara. He has curly black hair, pale-green eyes, and the relaxed demeanor of someone who lives five minutes away from his favorite beach. When he speaks, he tends to get distracted by his own digressions. He might begin with a point about memory, which reminds him of a favorite William James quote, which inspires a long soliloquy on the importance of introspection. Before long, we’re looking at pictures from Burning Man on his iPhone, which leads us back to the fragile nature of memory.
Although verbal overshadowing remains a widely accepted theory—it’s often invoked in the context of eyewitness testimony, for instance—Schooler is still a little peeved at the cosmos. “I know I should just move on already,” he says. “I really should stop talking about this. But I can’t.” That’s because he is convinced that he has stumbled on a serious problem, one that afflicts many of the most exciting new ideas in psychology.
One of the first demonstrations of this mysterious phenomenon came in the early nineteen-thirties. Joseph Banks Rhine, a psychologist at Duke, had developed an interest in the possibility of extrasensory perception, or E.S.P. Rhine devised an experiment featuring Zener cards, a special deck of twenty-five cards printed with one of five different symbols: a card was drawn from the deck and the subject was asked to guess the symbol. Most of Rhine’s subjects guessed about twenty per cent of the cards correctly, as you’d expect, but an undergraduate named Adam Linzmayer averaged nearly fifty per cent during his initial sessions, and pulled off several uncanny streaks, such as guessing nine cards in a row. The odds of this happening by chance are about one in two million. Linzmayer did it three times.
Rhine documented these stunning results in his notebook and prepared several papers for publication. But then, just as he began to believe in the possibility of extrasensory perception, the student lost his spooky talent. Between 1931 and 1933, Linzmayer guessed at the identity of another several thousand cards, but his success rate was now barely above chance. Rhine was forced to conclude that the student’s “extra-sensory perception ability has gone through a marked decline.” And Linzmayer wasn’t the only subject to experience such a drop-off: in nearly every case in which Rhine and others documented E.S.P. the effect dramatically diminished over time. Rhine called this trend the “decline effect.”
Schooler was fascinated by Rhine’s experimental struggles. Here was a scientist who had repeatedly documented the decline of his data; he seemed to have a talent for finding results that fell apart. In 2004, Schooler embarked on an ironic imitation of Rhine’s research: he tried to replicate this failure to replicate. In homage to Rhine’s interests, he decided to test for a parapsychological phenomenon known as precognition. The experiment itself was straightforward: he flashed a set of images to a subject and asked him or her to identify each one. Most of the time, the response was negative—the images were displayed too quickly to register. Then Schooler randomly selected half of the images to be shown again. What he wanted to know was whether the images that got a second showing were more likely to have been identified the first time around. Could subsequent exposure have somehow influenced the initial results? Could the effect become the cause?
from the issue
cartoon bank
e-mail this
The craziness of the hypothesis was the point: Schooler knows that precognition lacks a scientific explanation. But he wasn’t testing extrasensory powers; he was testing the decline effect. “At first, the data looked amazing, just as we’d expected,” Schooler says. “I couldn’t believe the amount of precognition we were finding. But then, as we kept on running subjects, the effect size”—a standard statistical measure—“kept on getting smaller and smaller.” The scientists eventually tested more than two thousand undergraduates. “In the end, our results looked just like Rhine’s,” Schooler said. “We found this strong paranormal effect, but it disappeared on us.”
The most likely explanation for the decline is an obvious one: regression to the mean. As the experiment is repeated, that is, an early statistical fluke gets cancelled out. The extrasensory powers of Schooler’s subjects didn’t decline—they were simply an illusion that vanished over time. And yet Schooler has noticed that many of the data sets that end up declining seem statistically solid—that is, they contain enough data that any regression to the mean shouldn’t be dramatic. “These are the results that pass all the tests,” he says. “The odds of them being random are typically quite remote, like one in a million. This means that the decline effect should almost never happen. But it happens all the time! Hell, it’s happened to me multiple times.” And this is why Schooler believes that the decline effect deserves more attention: its ubiquity seems to violate the laws of statistics. “Whenever I start talking about this, scientists get very nervous,” he says. “But I still want to know what happened to my results. Like most scientists, I assumed that it would get easier to document my effect over time. I’d get better at doing the experiments, at zeroing in on the conditions that produce verbal overshadowing. So why did the opposite happen? I’m convinced that we can use the tools of science to figure this out. First, though, we have to admit that we’ve got a problem.”
In 1991, the Danish zoologist Anders Møller, at Uppsala University, in Sweden, made a remarkable discovery about sex, barn swallows, and symmetry. It had long been known that the asymmetrical appearance of a creature was directly linked to the amount of mutation in its genome, so that more mutations led to more “fluctuating asymmetry.” (An easy way to measure asymmetry in humans is to compare the length of the fingers on each hand.) What Møller discovered is that female barn swallows were far more likely to mate with male birds that had long, symmetrical feathers. This suggested that the picky females were using symmetry as a proxy for the quality of male genes. Møller’s paper, which was published in Nature, set off a frenzy of research. Here was an easily measured, widely applicable indicator of genetic quality, and females could be shown to gravitate toward it. Aesthetics was really about genetics.
In the three years following, there were ten independent tests of the role of fluctuating asymmetry in sexual selection, and nine of them found a relationship between symmetry and male reproductive success. It didn’t matter if scientists were looking at the hairs on fruit flies or replicating the swallow studies—females seemed to prefer males with mirrored halves. Before long, the theory was applied to humans. Researchers found, for instance, that women preferred the smell of symmetrical men, but only during the fertile phase of the menstrual cycle. Other studies claimed that females had more orgasms when their partners were symmetrical, while a paper by anthropologists at Rutgers analyzed forty Jamaican dance routines and discovered that symmetrical men were consistently rated as better dancers.
Then the theory started to fall apart. In 1994, there were fourteen published tests of symmetry and sexual selection, and only eight found a correlation. In 1995, there were eight papers on the subject, and only four got a positive result. By 1998, when there were twelve additional investigations of fluctuating asymmetry, only a third of them confirmed the theory. Worse still, even the studies that yielded some positive result showed a steadily declining effect size. Between 1992 and 1997, the average effect size shrank by eighty per cent.
And it’s not just fluctuating asymmetry. In 2001, Michael Jennions, a biologist at the Australian National University, set out to analyze “temporal trends” across a wide range of subjects in ecology and evolutionary biology. He looked at hundreds of papers and forty-four meta-analyses (that is, statistical syntheses of related studies), and discovered a consistent decline effect over time, as many of the theories seemed to fade into irrelevance. In fact, even when numerous variables were controlled for—Jennions knew, for instance, that the same author might publish several critical papers, which could distort his analysis—there was still a significant decrease in the validity of the hypothesis, often within a year of publication. Jennions admits that his findings are troubling, but expresses a reluctance to talk about them publicly. “This is a very sensitive issue for scientists,” he says. “You know, we’re supposed to be dealing with hard facts, the stuff that’s supposed to stand the test of time. But when you see these trends you become a little more skeptical of things.”
What happened? Leigh Simmons, a biologist at the University of Western Australia, suggested one explanation when he told me about his initial enthusiasm for the theory: “I was really excited by fluctuating asymmetry. The early studies made the effect look very robust.” He decided to conduct a few experiments of his own, investigating symmetry in male horned beetles. “Unfortunately, I couldn’t find the effect,” he said. “But the worst part was that when I submitted these null results I had difficulty getting them published. The journals only wanted confirming data. It was too exciting an idea to disprove, at least back then.” For Simmons, the steep rise and slow fall of fluctuating asymmetry is a clear example of a scientific paradigm, one of those intellectual fads that both guide and constrain research: after a new paradigm is proposed, the peer-review process is tilted toward positive results. But then, after a few years, the academic incentives shift—the paradigm has become entrenched—so that the most notable results are now those that disprove the theory.
from the issue
cartoon bank
e-mail this
Jennions, similarly, argues that the decline effect is largely a product of publication bias, or the tendency of scientists and scientific journals to prefer positive data over null results, which is what happens when no effect is found. The bias was first identified by the statistician Theodore Sterling, in 1959, after he noticed that ninety-seven per cent of all published psychological studies with statistically significant data found the effect they were looking for. A “significant” result is defined as any data point that would be produced by chance less than five per cent of the time. This ubiquitous test was invented in 1922 by the English mathematician Ronald Fisher, who picked five per cent as the boundary line, somewhat arbitrarily, because it made pencil and slide-rule calculations easier. Sterling saw that if ninety-seven per cent of psychology studies were proving their hypotheses, either psychologists were extraordinarily lucky or they published only the outcomes of successful experiments. In recent years, publication bias has mostly been seen as a problem for clinical trials, since pharmaceutical companies are less interested in publishing results that aren’t favorable. But it’s becoming increasingly clear that publication bias also produces major distortions in fields without large corporate incentives, such as psychology and ecology.
While publication bias almost certainly plays a role in the decline effect, it remains an incomplete explanation. For one thing, it fails to account for the initial prevalence of positive results among studies that never even get submitted to journals. It also fails to explain the experience of people like Schooler, who have been unable to replicate their initial data despite their best efforts. Richard Palmer, a biologist at the University of Alberta, who has studied the problems surrounding fluctuating asymmetry, suspects that an equally significant issue is the selective reporting of results—the data that scientists choose to document in the first place. Palmer’s most convincing evidence relies on a statistical tool known as a funnel graph. When a large number of studies have been done on a single subject, the data should follow a pattern: studies with a large sample size should all cluster around a common value—the true result—whereas those with a smaller sample size should exhibit a random scattering, since they’re subject to greater sampling error. This pattern gives the graph its name, since the distribution resembles a funnel.
The funnel graph visually captures the distortions of selective reporting. For instance, after Palmer plotted every study of fluctuating asymmetry, he noticed that the distribution of results with smaller sample sizes wasn’t random at all but instead skewed heavily toward positive results. Palmer has since documented a similar problem in several other contested subject areas. “Once I realized that selective reporting is everywhere in science, I got quite depressed,” Palmer told me. “As a researcher, you’re always aware that there might be some nonrandom patterns, but I had no idea how widespread it is.” In a recent review article, Palmer summarized the impact of selective reporting on his field: “We cannot escape the troubling conclusion that some—perhaps many—cherished generalities are at best exaggerated in their biological significance and at worst a collective illusion nurtured by strong a-priori beliefs often repeated.”
Palmer emphasizes that selective reporting is not the same as scientific fraud. Rather, the problem seems to be one of subtle omissions and unconscious misperceptions, as researchers struggle to make sense of their results. Stephen Jay Gould referred to this as the “shoehorning” process. “A lot of scientific measurement is really hard,” Simmons told me. “If you’re talking about fluctuating asymmetry, then it’s a matter of minuscule differences between the right and left sides of an animal. It’s millimetres of a tail feather. And so maybe a researcher knows that he’s measuring a good male”—an animal that has successfully mated—“and he knows that it’s supposed to be symmetrical. Well, that act of measurement is going to be vulnerable to all sorts of perception biases. That’s not a cynical statement. That’s just the way human beings work.”
One of the classic examples of selective reporting concerns the testing of acupuncture in different countries. While acupuncture is widely accepted as a medical treatment in various Asian countries, its use is much more contested in the West. These cultural differences have profoundly influenced the results of clinical trials. Between 1966 and 1995, there were forty-seven studies of acupuncture in China, Taiwan, and Japan, and every single trial concluded that acupuncture was an effective treatment. During the same period, there were ninety-four clinical trials of acupuncture in the United States, Sweden, and the U.K., and only fifty-six per cent of these studies found any therapeutic benefits. As Palmer notes, this wide discrepancy suggests that scientists find ways to confirm their preferred hypothesis, disregarding what they don’t want to see. Our beliefs are a form of blindness.
John Ioannidis, an epidemiologist at Stanford University, argues that such distortions are a serious issue in biomedical research. “These exaggerations are why the decline has become so common,” he says. “It’d be really great if the initial studies gave us an accurate summary of things. But they don’t. And so what happens is we waste a lot of money treating millions of patients and doing lots of follow-up studies on other themes based on results that are misleading.” In 2005, Ioannidis published an article in the Journal of the American Medical Association that looked at the forty-nine most cited clinical-research studies in three major medical journals. Forty-five of these studies reported positive results, suggesting that the intervention being tested was effective. Because most of these studies were randomized controlled trials—the “gold standard” of medical evidence—they tended to have a significant impact on clinical practice, and led to the spread of treatments such as hormone replacement therapy for menopausal women and daily low-dose aspirin to prevent heart attacks and strokes. Nevertheless, the data Ioannidis found were disturbing: of the thirty-four claims that had been subject to replication, forty-one per cent had either been directly contradicted or had their effect sizes significantly downgraded.
The situation is even worse when a subject is fashionable. In recent years, for instance, there have been hundreds of studies on the various genes that control the differences in disease risk between men and women. These findings have included everything from the mutations responsible for the increased risk of schizophrenia to the genes underlying hypertension. Ioannidis and his colleagues looked at four hundred and thirty-two of these claims. They quickly discovered that the vast majority had serious flaws. But the most troubling fact emerged when he looked at the test of replication: out of four hundred and thirty-two claims, only a single one was consistently replicable. “This doesn’t mean that none of these claims will turn out to be true,” he says. “But, given that most of them were done badly, I wouldn’t hold my breath.”
from the issue
cartoon bank
e-mail this
According to Ioannidis, the main problem is that too many researchers engage in what he calls “significance chasing,” or finding ways to interpret the data so that it passes the statistical test of significance—the ninety-five-per-cent boundary invented by Ronald Fisher. “The scientists are so eager to pass this magical test that they start playing around with the numbers, trying to find anything that seems worthy,” Ioannidis says. In recent years, Ioannidis has become increasingly blunt about the pervasiveness of the problem. One of his most cited papers has a deliberately provocative title: “Why Most Published Research Findings Are False.”
The problem of selective reporting is rooted in a fundamental cognitive flaw, which is that we like proving ourselves right and hate being wrong. “It feels good to validate a hypothesis,” Ioannidis said. “It feels even better when you’ve got a financial interest in the idea or your career depends upon it. And that’s why, even after a claim has been systematically disproven”—he cites, for instance, the early work on hormone replacement therapy, or claims involving various vitamins—“you still see some stubborn researchers citing the first few studies that show a strong effect. They really want to believe that it’s true.”
That’s why Schooler argues that scientists need to become more rigorous about data collection before they publish. “We’re wasting too much time chasing after bad studies and underpowered experiments,” he says. The current “obsession” with replicability distracts from the real problem, which is faulty design. He notes that nobody even tries to replicate most science papers—there are simply too many. (According to Nature, a third of all studies never even get cited, let alone repeated.) “I’ve learned the hard way to be exceedingly careful,” Schooler says. “Every researcher should have to spell out, in advance, how many subjects they’re going to use, and what exactly they’re testing, and what constitutes a sufficient level of proof. We have the tools to be much more transparent about our experiments.”
In a forthcoming paper, Schooler recommends the establishment of an open-source database, in which researchers are required to outline their planned investigations and document all their results. “I think this would provide a huge increase in access to scientific work and give us a much better way to judge the quality of an experiment,” Schooler says. “It would help us finally deal with all these issues that the decline effect is exposing.”
Although such reforms would mitigate the dangers of publication bias and selective reporting, they still wouldn’t erase the decline effect. This is largely because scientific research will always be shadowed by a force that can’t be curbed, only contained: sheer randomness. Although little research has been done on the experimental dangers of chance and happenstance, the research that exists isn’t encouraging.
In the late nineteen-nineties, John Crabbe, a neuroscientist at the Oregon Health and Science University, conducted an experiment that showed how unknowable chance events can skew tests of replicability. He performed a series of experiments on mouse behavior in three different science labs: in Albany, New York; Edmonton, Alberta; and Portland, Oregon. Before he conducted the experiments, he tried to standardize every variable he could think of. The same strains of mice were used in each lab, shipped on the same day from the same supplier. The animals were raised in the same kind of enclosure, with the same brand of sawdust bedding. They had been exposed to the same amount of incandescent light, were living with the same number of littermates, and were fed the exact same type of chow pellets. When the mice were handled, it was with the same kind of surgical glove, and when they were tested it was on the same equipment, at the same time in the morning.
The premise of this test of replicability, of course, is that each of the labs should have generated the same pattern of results. “If any set of experiments should have passed the test, it should have been ours,” Crabbe says. “But that’s not the way it turned out.” In one experiment, Crabbe injected a particular strain of mouse with cocaine. In Portland the mice given the drug moved, on average, six hundred centimetres more than they normally did; in Albany they moved seven hundred and one additional centimetres. But in the Edmonton lab they moved more than five thousand additional centimetres. Similar deviations were observed in a test of anxiety. Furthermore, these inconsistencies didn’t follow any detectable pattern. In Portland one strain of mouse proved most anxious, while in Albany another strain won that distinction.
The disturbing implication of the Crabbe study is that a lot of extraordinary scientific data are nothing but noise. The hyperactivity of those coked-up Edmonton mice wasn’t an interesting new fact—it was a meaningless outlier, a by-product of invisible variables we don’t understand. The problem, of course, is that such dramatic findings are also the most likely to get published in prestigious journals, since the data are both statistically significant and entirely unexpected. Grants get written, follow-up studies are conducted. The end result is a scientific accident that can take years to unravel.
from the issue
cartoon bank
e-mail this
This suggests that the decline effect is actually a decline of illusion. While Karl Popper imagined falsification occurring with a single, definitive experiment—Galileo refuted Aristotelian mechanics in an afternoon—the process turns out to be much messier than that. Many scientific theories continue to be considered true even after failing numerous experimental tests. Verbal overshadowing might exhibit the decline effect, but it remains extensively relied upon within the field. The same holds for any number of phenomena, from the disappearing benefits of second-generation antipsychotics to the weak coupling ratio exhibited by decaying neutrons, which appears to have fallen by more than ten standard deviations between 1969 and 2001. Even the law of gravity hasn’t always been perfect at predicting real-world phenomena. (In one test, physicists measuring gravity by means of deep boreholes in the Nevada desert found a two-and-a-half-per-cent discrepancy between the theoretical predictions and the actual data.) Despite these findings, second-generation antipsychotics are still widely prescribed, and our model of the neutron hasn’t changed. The law of gravity remains the same.
Such anomalies demonstrate the slipperiness of empiricism. Although many scientific ideas generate conflicting results and suffer from falling effect sizes, they continue to get cited in the textbooks and drive standard medical practice. Why? Because these ideas seem true. Because they make sense. Because we can’t bear to let them go. And this is why the decline effect is so troubling. Not because it reveals the human fallibility of science, in which data are tweaked and beliefs shape perceptions. (Such shortcomings aren’t surprising, at least for scientists.) And not because it reveals that many of our most exciting theories are fleeting fads and will soon be rejected. (That idea has been around since Thomas Kuhn.) The decline effect is troubling because it reminds us how difficult it is to prove anything. We like to pretend that our experiments define the truth for us. But that’s often not the case. Just because an idea is true doesn’t mean it can be proved. And just because an idea can be proved doesn’t mean it’s true. When the experiments are done, we still have to choose what to believe. ♦
Read more http://www.newyorker.com/reporting/2010/12/13/101213fa_fact_lehrer#ixzz1BlB1sWp0
Is there something wrong with the scientific method?by Jonah Lehrer
December 13, 2010
Read more http://www.newyorker.com/reporting/2010/12/13/101213fa_fact_lehrer#ixzz1Bl8DSHE5
The Truth Wears OffIs there something wrong with the scientific method?by Jonah Lehrer
December 13, 2010 Many results that are rigorously proved and accepted start shrinking in later studies.
Share Print E-Mail Single Page Keywords
Scientific Experiments; Decline Effect; Replicability; Scientists; Statistics; Jonathan Schooler; Scientific Theories On September 18, 2007, a few dozen neuroscientists, psychiatrists, and drug-company executives gathered in a hotel conference room in Brussels to hear some startling news. It had to do with a class of drugs known as atypical or second-generation antipsychotics, which came on the market in the early nineties. The drugs, sold under brand names such as Abilify, Seroquel, and Zyprexa, had been tested on schizophrenics in several large clinical trials, all of which had demonstrated a dramatic decrease in the subjects’ psychiatric symptoms. As a result, second-generation antipsychotics had become one of the fastest-growing and most profitable pharmaceutical classes. By 2001, Eli Lilly’s Zyprexa was generating more revenue than Prozac. It remains the company’s top-selling drug.
But the data presented at the Brussels meeting made it clear that something strange was happening: the therapeutic power of the drugs appeared to be steadily waning. A recent study showed an effect that was less than half of that documented in the first trials, in the early nineteen-nineties. Many researchers began to argue that the expensive pharmaceuticals weren’t any better than first-generation antipsychotics, which have been in use since the fifties. “In fact, sometimes they now look even worse,” John Davis, a professor of psychiatry at the University of Illinois at Chicago, told me.
Before the effectiveness of a drug can be confirmed, it must be tested and tested again. Different scientists in different labs need to repeat the protocols and publish their results. The test of replicability, as it’s known, is the foundation of modern research. Replicability is how the community enforces itself. It’s a safeguard for the creep of subjectivity. Most of the time, scientists know what results they want, and that can influence the results they get. The premise of replicability is that the scientific community can correct for these flaws.
But now all sorts of well-established, multiply confirmed findings have started to look increasingly uncertain. It’s as if our facts were losing their truth: claims that have been enshrined in textbooks are suddenly unprovable. This phenomenon doesn’t yet have an official name, but it’s occurring across a wide range of fields, from psychology to ecology. In the field of medicine, the phenomenon seems extremely widespread, affecting not only antipsychotics but also therapies ranging from cardiac stents to Vitamin E and antidepressants: Davis has a forthcoming analysis demonstrating that the efficacy of antidepressants has gone down as much as threefold in recent decades.
For many scientists, the effect is especially troubling because of what it exposes about the scientific process. If replication is what separates the rigor of science from the squishiness of pseudoscience, where do we put all these rigorously validated findings that can no longer be proved? Which results should we believe? Francis Bacon, the early-modern philosopher and pioneer of the scientific method, once declared that experiments were essential, because they allowed us to “put nature to the question.” But it appears that nature often gives us different answers.
from the issuecartoon banke-mail thisJonathan Schooler was a young graduate student at the University of Washington in the nineteen-eighties when he discovered a surprising new fact about language and memory. At the time, it was widely believed that the act of describing our memories improved them. But, in a series of clever experiments, Schooler demonstrated that subjects shown a face and asked to describe it were much less likely to recognize the face when shown it later than those who had simply looked at it. Schooler called the phenomenon “verbal overshadowing.”
The study turned him into an academic star. Since its initial publication, in 1990, it has been cited more than four hundred times. Before long, Schooler had extended the model to a variety of other tasks, such as remembering the taste of a wine, identifying the best strawberry jam, and solving difficult creative puzzles. In each instance, asking people to put their perceptions into words led to dramatic decreases in performance.
But while Schooler was publishing these results in highly reputable journals, a secret worry gnawed at him: it was proving difficult to replicate his earlier findings. “I’d often still see an effect, but the effect just wouldn’t be as strong,” he told me. “It was as if verbal overshadowing, my big new idea, was getting weaker.” At first, he assumed that he’d made an error in experimental design or a statistical miscalculation. But he couldn’t find anything wrong with his research. He then concluded that his initial batch of research subjects must have been unusually susceptible to verbal overshadowing. (John Davis, similarly, has speculated that part of the drop-off in the effectiveness of antipsychotics can be attributed to using subjects who suffer from milder forms of psychosis which are less likely to show dramatic improvement.) “It wasn’t a very satisfying explanation,” Schooler says. “One of my mentors told me that my real mistake was trying to replicate my work. He told me doing that was just setting myself up for disappointment.”
Schooler tried to put the problem out of his mind; his colleagues assured him that such things happened all the time. Over the next few years, he found new research questions, got married and had kids. But his replication problem kept on getting worse. His first attempt at replicating the 1990 study, in 1995, resulted in an effect that was thirty per cent smaller. The next year, the size of the effect shrank another thirty per cent. When other labs repeated Schooler’s experiments, they got a similar spread of data, with a distinct downward trend. “This was profoundly frustrating,” he says. “It was as if nature gave me this great result and then tried to take it back.” In private, Schooler began referring to the problem as “cosmic habituation,” by analogy to the decrease in response that occurs when individuals habituate to particular stimuli. “Habituation is why you don’t notice the stuff that’s always there,” Schooler says. “It’s an inevitable process of adjustment, a ratcheting down of excitement. I started joking that it was like the cosmos was habituating to my ideas. I took it very personally.”
Schooler is now a tenured professor at the University of California at Santa Barbara. He has curly black hair, pale-green eyes, and the relaxed demeanor of someone who lives five minutes away from his favorite beach. When he speaks, he tends to get distracted by his own digressions. He might begin with a point about memory, which reminds him of a favorite William James quote, which inspires a long soliloquy on the importance of introspection. Before long, we’re looking at pictures from Burning Man on his iPhone, which leads us back to the fragile nature of memory.
Although verbal overshadowing remains a widely accepted theory—it’s often invoked in the context of eyewitness testimony, for instance—Schooler is still a little peeved at the cosmos. “I know I should just move on already,” he says. “I really should stop talking about this. But I can’t.” That’s because he is convinced that he has stumbled on a serious problem, one that afflicts many of the most exciting new ideas in psychology.
One of the first demonstrations of this mysterious phenomenon came in the early nineteen-thirties. Joseph Banks Rhine, a psychologist at Duke, had developed an interest in the possibility of extrasensory perception, or E.S.P. Rhine devised an experiment featuring Zener cards, a special deck of twenty-five cards printed with one of five different symbols: a card was drawn from the deck and the subject was asked to guess the symbol. Most of Rhine’s subjects guessed about twenty per cent of the cards correctly, as you’d expect, but an undergraduate named Adam Linzmayer averaged nearly fifty per cent during his initial sessions, and pulled off several uncanny streaks, such as guessing nine cards in a row. The odds of this happening by chance are about one in two million. Linzmayer did it three times.
Rhine documented these stunning results in his notebook and prepared several papers for publication. But then, just as he began to believe in the possibility of extrasensory perception, the student lost his spooky talent. Between 1931 and 1933, Linzmayer guessed at the identity of another several thousand cards, but his success rate was now barely above chance. Rhine was forced to conclude that the student’s “extra-sensory perception ability has gone through a marked decline.” And Linzmayer wasn’t the only subject to experience such a drop-off: in nearly every case in which Rhine and others documented E.S.P. the effect dramatically diminished over time. Rhine called this trend the “decline effect.”
Schooler was fascinated by Rhine’s experimental struggles. Here was a scientist who had repeatedly documented the decline of his data; he seemed to have a talent for finding results that fell apart. In 2004, Schooler embarked on an ironic imitation of Rhine’s research: he tried to replicate this failure to replicate. In homage to Rhine’s interests, he decided to test for a parapsychological phenomenon known as precognition. The experiment itself was straightforward: he flashed a set of images to a subject and asked him or her to identify each one. Most of the time, the response was negative—the images were displayed too quickly to register. Then Schooler randomly selected half of the images to be shown again. What he wanted to know was whether the images that got a second showing were more likely to have been identified the first time around. Could subsequent exposure have somehow influenced the initial results? Could the effect become the cause?
from the issue
cartoon bank
e-mail this
The craziness of the hypothesis was the point: Schooler knows that precognition lacks a scientific explanation. But he wasn’t testing extrasensory powers; he was testing the decline effect. “At first, the data looked amazing, just as we’d expected,” Schooler says. “I couldn’t believe the amount of precognition we were finding. But then, as we kept on running subjects, the effect size”—a standard statistical measure—“kept on getting smaller and smaller.” The scientists eventually tested more than two thousand undergraduates. “In the end, our results looked just like Rhine’s,” Schooler said. “We found this strong paranormal effect, but it disappeared on us.”
The most likely explanation for the decline is an obvious one: regression to the mean. As the experiment is repeated, that is, an early statistical fluke gets cancelled out. The extrasensory powers of Schooler’s subjects didn’t decline—they were simply an illusion that vanished over time. And yet Schooler has noticed that many of the data sets that end up declining seem statistically solid—that is, they contain enough data that any regression to the mean shouldn’t be dramatic. “These are the results that pass all the tests,” he says. “The odds of them being random are typically quite remote, like one in a million. This means that the decline effect should almost never happen. But it happens all the time! Hell, it’s happened to me multiple times.” And this is why Schooler believes that the decline effect deserves more attention: its ubiquity seems to violate the laws of statistics. “Whenever I start talking about this, scientists get very nervous,” he says. “But I still want to know what happened to my results. Like most scientists, I assumed that it would get easier to document my effect over time. I’d get better at doing the experiments, at zeroing in on the conditions that produce verbal overshadowing. So why did the opposite happen? I’m convinced that we can use the tools of science to figure this out. First, though, we have to admit that we’ve got a problem.”
In 1991, the Danish zoologist Anders Møller, at Uppsala University, in Sweden, made a remarkable discovery about sex, barn swallows, and symmetry. It had long been known that the asymmetrical appearance of a creature was directly linked to the amount of mutation in its genome, so that more mutations led to more “fluctuating asymmetry.” (An easy way to measure asymmetry in humans is to compare the length of the fingers on each hand.) What Møller discovered is that female barn swallows were far more likely to mate with male birds that had long, symmetrical feathers. This suggested that the picky females were using symmetry as a proxy for the quality of male genes. Møller’s paper, which was published in Nature, set off a frenzy of research. Here was an easily measured, widely applicable indicator of genetic quality, and females could be shown to gravitate toward it. Aesthetics was really about genetics.
In the three years following, there were ten independent tests of the role of fluctuating asymmetry in sexual selection, and nine of them found a relationship between symmetry and male reproductive success. It didn’t matter if scientists were looking at the hairs on fruit flies or replicating the swallow studies—females seemed to prefer males with mirrored halves. Before long, the theory was applied to humans. Researchers found, for instance, that women preferred the smell of symmetrical men, but only during the fertile phase of the menstrual cycle. Other studies claimed that females had more orgasms when their partners were symmetrical, while a paper by anthropologists at Rutgers analyzed forty Jamaican dance routines and discovered that symmetrical men were consistently rated as better dancers.
Then the theory started to fall apart. In 1994, there were fourteen published tests of symmetry and sexual selection, and only eight found a correlation. In 1995, there were eight papers on the subject, and only four got a positive result. By 1998, when there were twelve additional investigations of fluctuating asymmetry, only a third of them confirmed the theory. Worse still, even the studies that yielded some positive result showed a steadily declining effect size. Between 1992 and 1997, the average effect size shrank by eighty per cent.
And it’s not just fluctuating asymmetry. In 2001, Michael Jennions, a biologist at the Australian National University, set out to analyze “temporal trends” across a wide range of subjects in ecology and evolutionary biology. He looked at hundreds of papers and forty-four meta-analyses (that is, statistical syntheses of related studies), and discovered a consistent decline effect over time, as many of the theories seemed to fade into irrelevance. In fact, even when numerous variables were controlled for—Jennions knew, for instance, that the same author might publish several critical papers, which could distort his analysis—there was still a significant decrease in the validity of the hypothesis, often within a year of publication. Jennions admits that his findings are troubling, but expresses a reluctance to talk about them publicly. “This is a very sensitive issue for scientists,” he says. “You know, we’re supposed to be dealing with hard facts, the stuff that’s supposed to stand the test of time. But when you see these trends you become a little more skeptical of things.”
What happened? Leigh Simmons, a biologist at the University of Western Australia, suggested one explanation when he told me about his initial enthusiasm for the theory: “I was really excited by fluctuating asymmetry. The early studies made the effect look very robust.” He decided to conduct a few experiments of his own, investigating symmetry in male horned beetles. “Unfortunately, I couldn’t find the effect,” he said. “But the worst part was that when I submitted these null results I had difficulty getting them published. The journals only wanted confirming data. It was too exciting an idea to disprove, at least back then.” For Simmons, the steep rise and slow fall of fluctuating asymmetry is a clear example of a scientific paradigm, one of those intellectual fads that both guide and constrain research: after a new paradigm is proposed, the peer-review process is tilted toward positive results. But then, after a few years, the academic incentives shift—the paradigm has become entrenched—so that the most notable results are now those that disprove the theory.
from the issue
cartoon bank
e-mail this
Jennions, similarly, argues that the decline effect is largely a product of publication bias, or the tendency of scientists and scientific journals to prefer positive data over null results, which is what happens when no effect is found. The bias was first identified by the statistician Theodore Sterling, in 1959, after he noticed that ninety-seven per cent of all published psychological studies with statistically significant data found the effect they were looking for. A “significant” result is defined as any data point that would be produced by chance less than five per cent of the time. This ubiquitous test was invented in 1922 by the English mathematician Ronald Fisher, who picked five per cent as the boundary line, somewhat arbitrarily, because it made pencil and slide-rule calculations easier. Sterling saw that if ninety-seven per cent of psychology studies were proving their hypotheses, either psychologists were extraordinarily lucky or they published only the outcomes of successful experiments. In recent years, publication bias has mostly been seen as a problem for clinical trials, since pharmaceutical companies are less interested in publishing results that aren’t favorable. But it’s becoming increasingly clear that publication bias also produces major distortions in fields without large corporate incentives, such as psychology and ecology.
While publication bias almost certainly plays a role in the decline effect, it remains an incomplete explanation. For one thing, it fails to account for the initial prevalence of positive results among studies that never even get submitted to journals. It also fails to explain the experience of people like Schooler, who have been unable to replicate their initial data despite their best efforts. Richard Palmer, a biologist at the University of Alberta, who has studied the problems surrounding fluctuating asymmetry, suspects that an equally significant issue is the selective reporting of results—the data that scientists choose to document in the first place. Palmer’s most convincing evidence relies on a statistical tool known as a funnel graph. When a large number of studies have been done on a single subject, the data should follow a pattern: studies with a large sample size should all cluster around a common value—the true result—whereas those with a smaller sample size should exhibit a random scattering, since they’re subject to greater sampling error. This pattern gives the graph its name, since the distribution resembles a funnel.
The funnel graph visually captures the distortions of selective reporting. For instance, after Palmer plotted every study of fluctuating asymmetry, he noticed that the distribution of results with smaller sample sizes wasn’t random at all but instead skewed heavily toward positive results. Palmer has since documented a similar problem in several other contested subject areas. “Once I realized that selective reporting is everywhere in science, I got quite depressed,” Palmer told me. “As a researcher, you’re always aware that there might be some nonrandom patterns, but I had no idea how widespread it is.” In a recent review article, Palmer summarized the impact of selective reporting on his field: “We cannot escape the troubling conclusion that some—perhaps many—cherished generalities are at best exaggerated in their biological significance and at worst a collective illusion nurtured by strong a-priori beliefs often repeated.”
Palmer emphasizes that selective reporting is not the same as scientific fraud. Rather, the problem seems to be one of subtle omissions and unconscious misperceptions, as researchers struggle to make sense of their results. Stephen Jay Gould referred to this as the “shoehorning” process. “A lot of scientific measurement is really hard,” Simmons told me. “If you’re talking about fluctuating asymmetry, then it’s a matter of minuscule differences between the right and left sides of an animal. It’s millimetres of a tail feather. And so maybe a researcher knows that he’s measuring a good male”—an animal that has successfully mated—“and he knows that it’s supposed to be symmetrical. Well, that act of measurement is going to be vulnerable to all sorts of perception biases. That’s not a cynical statement. That’s just the way human beings work.”
One of the classic examples of selective reporting concerns the testing of acupuncture in different countries. While acupuncture is widely accepted as a medical treatment in various Asian countries, its use is much more contested in the West. These cultural differences have profoundly influenced the results of clinical trials. Between 1966 and 1995, there were forty-seven studies of acupuncture in China, Taiwan, and Japan, and every single trial concluded that acupuncture was an effective treatment. During the same period, there were ninety-four clinical trials of acupuncture in the United States, Sweden, and the U.K., and only fifty-six per cent of these studies found any therapeutic benefits. As Palmer notes, this wide discrepancy suggests that scientists find ways to confirm their preferred hypothesis, disregarding what they don’t want to see. Our beliefs are a form of blindness.
John Ioannidis, an epidemiologist at Stanford University, argues that such distortions are a serious issue in biomedical research. “These exaggerations are why the decline has become so common,” he says. “It’d be really great if the initial studies gave us an accurate summary of things. But they don’t. And so what happens is we waste a lot of money treating millions of patients and doing lots of follow-up studies on other themes based on results that are misleading.” In 2005, Ioannidis published an article in the Journal of the American Medical Association that looked at the forty-nine most cited clinical-research studies in three major medical journals. Forty-five of these studies reported positive results, suggesting that the intervention being tested was effective. Because most of these studies were randomized controlled trials—the “gold standard” of medical evidence—they tended to have a significant impact on clinical practice, and led to the spread of treatments such as hormone replacement therapy for menopausal women and daily low-dose aspirin to prevent heart attacks and strokes. Nevertheless, the data Ioannidis found were disturbing: of the thirty-four claims that had been subject to replication, forty-one per cent had either been directly contradicted or had their effect sizes significantly downgraded.
The situation is even worse when a subject is fashionable. In recent years, for instance, there have been hundreds of studies on the various genes that control the differences in disease risk between men and women. These findings have included everything from the mutations responsible for the increased risk of schizophrenia to the genes underlying hypertension. Ioannidis and his colleagues looked at four hundred and thirty-two of these claims. They quickly discovered that the vast majority had serious flaws. But the most troubling fact emerged when he looked at the test of replication: out of four hundred and thirty-two claims, only a single one was consistently replicable. “This doesn’t mean that none of these claims will turn out to be true,” he says. “But, given that most of them were done badly, I wouldn’t hold my breath.”
from the issue
cartoon bank
e-mail this
According to Ioannidis, the main problem is that too many researchers engage in what he calls “significance chasing,” or finding ways to interpret the data so that it passes the statistical test of significance—the ninety-five-per-cent boundary invented by Ronald Fisher. “The scientists are so eager to pass this magical test that they start playing around with the numbers, trying to find anything that seems worthy,” Ioannidis says. In recent years, Ioannidis has become increasingly blunt about the pervasiveness of the problem. One of his most cited papers has a deliberately provocative title: “Why Most Published Research Findings Are False.”
The problem of selective reporting is rooted in a fundamental cognitive flaw, which is that we like proving ourselves right and hate being wrong. “It feels good to validate a hypothesis,” Ioannidis said. “It feels even better when you’ve got a financial interest in the idea or your career depends upon it. And that’s why, even after a claim has been systematically disproven”—he cites, for instance, the early work on hormone replacement therapy, or claims involving various vitamins—“you still see some stubborn researchers citing the first few studies that show a strong effect. They really want to believe that it’s true.”
That’s why Schooler argues that scientists need to become more rigorous about data collection before they publish. “We’re wasting too much time chasing after bad studies and underpowered experiments,” he says. The current “obsession” with replicability distracts from the real problem, which is faulty design. He notes that nobody even tries to replicate most science papers—there are simply too many. (According to Nature, a third of all studies never even get cited, let alone repeated.) “I’ve learned the hard way to be exceedingly careful,” Schooler says. “Every researcher should have to spell out, in advance, how many subjects they’re going to use, and what exactly they’re testing, and what constitutes a sufficient level of proof. We have the tools to be much more transparent about our experiments.”
In a forthcoming paper, Schooler recommends the establishment of an open-source database, in which researchers are required to outline their planned investigations and document all their results. “I think this would provide a huge increase in access to scientific work and give us a much better way to judge the quality of an experiment,” Schooler says. “It would help us finally deal with all these issues that the decline effect is exposing.”
Although such reforms would mitigate the dangers of publication bias and selective reporting, they still wouldn’t erase the decline effect. This is largely because scientific research will always be shadowed by a force that can’t be curbed, only contained: sheer randomness. Although little research has been done on the experimental dangers of chance and happenstance, the research that exists isn’t encouraging.
In the late nineteen-nineties, John Crabbe, a neuroscientist at the Oregon Health and Science University, conducted an experiment that showed how unknowable chance events can skew tests of replicability. He performed a series of experiments on mouse behavior in three different science labs: in Albany, New York; Edmonton, Alberta; and Portland, Oregon. Before he conducted the experiments, he tried to standardize every variable he could think of. The same strains of mice were used in each lab, shipped on the same day from the same supplier. The animals were raised in the same kind of enclosure, with the same brand of sawdust bedding. They had been exposed to the same amount of incandescent light, were living with the same number of littermates, and were fed the exact same type of chow pellets. When the mice were handled, it was with the same kind of surgical glove, and when they were tested it was on the same equipment, at the same time in the morning.
The premise of this test of replicability, of course, is that each of the labs should have generated the same pattern of results. “If any set of experiments should have passed the test, it should have been ours,” Crabbe says. “But that’s not the way it turned out.” In one experiment, Crabbe injected a particular strain of mouse with cocaine. In Portland the mice given the drug moved, on average, six hundred centimetres more than they normally did; in Albany they moved seven hundred and one additional centimetres. But in the Edmonton lab they moved more than five thousand additional centimetres. Similar deviations were observed in a test of anxiety. Furthermore, these inconsistencies didn’t follow any detectable pattern. In Portland one strain of mouse proved most anxious, while in Albany another strain won that distinction.
The disturbing implication of the Crabbe study is that a lot of extraordinary scientific data are nothing but noise. The hyperactivity of those coked-up Edmonton mice wasn’t an interesting new fact—it was a meaningless outlier, a by-product of invisible variables we don’t understand. The problem, of course, is that such dramatic findings are also the most likely to get published in prestigious journals, since the data are both statistically significant and entirely unexpected. Grants get written, follow-up studies are conducted. The end result is a scientific accident that can take years to unravel.
from the issue
cartoon bank
e-mail this
This suggests that the decline effect is actually a decline of illusion. While Karl Popper imagined falsification occurring with a single, definitive experiment—Galileo refuted Aristotelian mechanics in an afternoon—the process turns out to be much messier than that. Many scientific theories continue to be considered true even after failing numerous experimental tests. Verbal overshadowing might exhibit the decline effect, but it remains extensively relied upon within the field. The same holds for any number of phenomena, from the disappearing benefits of second-generation antipsychotics to the weak coupling ratio exhibited by decaying neutrons, which appears to have fallen by more than ten standard deviations between 1969 and 2001. Even the law of gravity hasn’t always been perfect at predicting real-world phenomena. (In one test, physicists measuring gravity by means of deep boreholes in the Nevada desert found a two-and-a-half-per-cent discrepancy between the theoretical predictions and the actual data.) Despite these findings, second-generation antipsychotics are still widely prescribed, and our model of the neutron hasn’t changed. The law of gravity remains the same.
Such anomalies demonstrate the slipperiness of empiricism. Although many scientific ideas generate conflicting results and suffer from falling effect sizes, they continue to get cited in the textbooks and drive standard medical practice. Why? Because these ideas seem true. Because they make sense. Because we can’t bear to let them go. And this is why the decline effect is so troubling. Not because it reveals the human fallibility of science, in which data are tweaked and beliefs shape perceptions. (Such shortcomings aren’t surprising, at least for scientists.) And not because it reveals that many of our most exciting theories are fleeting fads and will soon be rejected. (That idea has been around since Thomas Kuhn.) The decline effect is troubling because it reminds us how difficult it is to prove anything. We like to pretend that our experiments define the truth for us. But that’s often not the case. Just because an idea is true doesn’t mean it can be proved. And just because an idea can be proved doesn’t mean it’s true. When the experiments are done, we still have to choose what to believe. ♦
Read more http://www.newyorker.com/reporting/2010/12/13/101213fa_fact_lehrer#ixzz1BlB1sWp0
среда, 8 декабря 2010 г.
Как глобальные корпорации стимулируют рост национальных стартапов
00:03 РБК daily
Центральная тема в разговорах о модернизации российской экономики — спрос на инновации, предъявляемый крупным бизнесом. В частности, нередко говорится, что технологический спрос со стороны больших корпораций является причиной рождения множества малых стартапов. Это не совсем точно — малые инновационные компании рождаются по другим причинам. Однако «большие» и «маленькие» являются элементами одной системы, которая устанавливает между ними тесные взаимосвязи. Если эта система связей сформирована — экономика развивается по инновационному сценарию, если нет — число успешных стартапов остается минимальным.
Условие развития любой малой компании — неординарный подход. Или продукт необычен, или создана новаторская бизнес-модель, или система продаж — нетрадиционна. Иными словами, у малой компании должно быть новое конкурентное преимущество — в противном случае она не выйдет на рынок, плотно занятый другими. Допустим, я хочу выйти на рынок бензоколонок. Это категорически невозможно, если я не предложу инновационного решения — например, создам технологию, при которой человеку не нужно выходить из машины. Архаичный вариант — люди заливают вам топливо, инновационный вариант — машина подъезжает, пистолет соединяется с баком, а оплата осуществляется с мобильного телефона. Остается вопрос с ценой: сколько готов заплатить потребитель за дополнительное удобство? Если рынок принимает новый продукт, и я очень точно, определяю оптимальную премию за новизну — у меня появляется шанс выиграть. Таким образом, причина рождения стартапов — высокая конкуренция. Большие корпорации уже заняли рынок, а малым еще предстоит это сделать. Флеш-карты, Windows Media Player, CD-ROM-привод — тысячи продуктов родились в стартапах. Но мы как потребители познакомились с ними через большие корпорации. Почему же большой бизнес поставил на поток продукты, изобретенные в стартапах?
Существует всем известный (и желанный многими) путь развития стартапа. Стартап, как мы рассмотрели на примере бензоколонок, разрабатывает востребованную рынком технологию и предлагает потребителю продукт этой технологии по «правильной» цене — но это лишь предпосылки для успеха. Дальше вам надо как можно быстрее выходить к инвестору, получать от него деньги, чтобы опять же максимально быстро выскочить на рынок и занять на нем максимально широкую нишу по объему продаж. Все это делается, в общем, с одной единственной целью — верхняя строка отчета о прибылях и убытках. Инвестиционное сообщество на основании этой строки, собственно, и определяет будущую стоимость акций (я немного утрирую, но в целом это так). Таким образом, добившись высокого объема продаж, стартап готов для проведения IPO по максимально высокой цене — отличный exit (выход из инвестиций) для любого инвестора. Классическая модель, только в ней очень много «если». Если венчурный инвестор вовремя не найдется, или не получится быстро достичь достаточного объема продаж, то пока вы пробиваете железные двери, крупная компания может наладить серийное производство, оставив вас вне игры. Но давайте взглянем на ситуацию по-другому.
Необязательно создавать новый рынок или новую нишу на традиционном — можно встраиваться в ниши, созданные другими. Каждая крупная корпорация создает новые ниши десятками, если не сотнями. Преимущество большой компании не в том, что она инновационна, а в том, что она обеспечена каналами продаж. Я хорошо знаком с бизнесом одной большой компании — знаю, в частности, что она аккумулировала cash в размере 16 млрд долл. Компания занимается высокотехнологичным производством, развивает НИОКР. Так почему бы, казалось бы, не вложить эти деньги в исследования и разработки? Проблема в том, что как только компания заявит о таком вложении, цена на ее акции существенно упадет, ведь НИОКР — вещь рискованная: то ли получится, то ли нет. Таким образом, все крупные корпорации (и это не преувеличение) «сидят» на наличных располагая очень ограниченным набором решений, как этими деньгами распорядиться. Возвращаясь к компании, которую я упомянул в связи с 16 млрд, отмечу, что ее бюджет на собственный НИОКР составляет порядка 2 млрд долл. и примерно в пять раз больше вкладывается в покупку стартапов. И любой крупный высокотехнологичный бизнес тратит на поиск подходящих для покупки активов колоссальное время, силы и средства (Google, без сомнения, станет крупнейшим игроком M&A в 2010 году).
Я уверен, что сотрудничество с большими корпорациями — естественный путь развития российских стартапов. Задача бизнес-инкубаторов и технопарков, как я ее вижу, заключается в том, чтобы корпорации четко сформулировали, в чем заключаются их потребности в области инноваций, и донесли эту информацию до стартапов. А с другой стороны, нужно найти стартапы, способные удовлетворить спрос большого бизнеса, и вырастить их до того состояния, когда они смогут с этой задачей успешно справиться. На глобальном рынке все корпорации заявляют об интересе к инновационным стартапам — на рынке России им еще предстоит этот интерес продекларировать.
Илья Толстов, генеральный директор технопарка «Ингрия»
Читать полностью: http://www.rbcdaily.ru/2010/12/08/focus/562949979263403.shtml
http://www.rbcdaily.ru/2010/12/08/focus/562949979263403
Центральная тема в разговорах о модернизации российской экономики — спрос на инновации, предъявляемый крупным бизнесом. В частности, нередко говорится, что технологический спрос со стороны больших корпораций является причиной рождения множества малых стартапов. Это не совсем точно — малые инновационные компании рождаются по другим причинам. Однако «большие» и «маленькие» являются элементами одной системы, которая устанавливает между ними тесные взаимосвязи. Если эта система связей сформирована — экономика развивается по инновационному сценарию, если нет — число успешных стартапов остается минимальным.
Условие развития любой малой компании — неординарный подход. Или продукт необычен, или создана новаторская бизнес-модель, или система продаж — нетрадиционна. Иными словами, у малой компании должно быть новое конкурентное преимущество — в противном случае она не выйдет на рынок, плотно занятый другими. Допустим, я хочу выйти на рынок бензоколонок. Это категорически невозможно, если я не предложу инновационного решения — например, создам технологию, при которой человеку не нужно выходить из машины. Архаичный вариант — люди заливают вам топливо, инновационный вариант — машина подъезжает, пистолет соединяется с баком, а оплата осуществляется с мобильного телефона. Остается вопрос с ценой: сколько готов заплатить потребитель за дополнительное удобство? Если рынок принимает новый продукт, и я очень точно, определяю оптимальную премию за новизну — у меня появляется шанс выиграть. Таким образом, причина рождения стартапов — высокая конкуренция. Большие корпорации уже заняли рынок, а малым еще предстоит это сделать. Флеш-карты, Windows Media Player, CD-ROM-привод — тысячи продуктов родились в стартапах. Но мы как потребители познакомились с ними через большие корпорации. Почему же большой бизнес поставил на поток продукты, изобретенные в стартапах?
Существует всем известный (и желанный многими) путь развития стартапа. Стартап, как мы рассмотрели на примере бензоколонок, разрабатывает востребованную рынком технологию и предлагает потребителю продукт этой технологии по «правильной» цене — но это лишь предпосылки для успеха. Дальше вам надо как можно быстрее выходить к инвестору, получать от него деньги, чтобы опять же максимально быстро выскочить на рынок и занять на нем максимально широкую нишу по объему продаж. Все это делается, в общем, с одной единственной целью — верхняя строка отчета о прибылях и убытках. Инвестиционное сообщество на основании этой строки, собственно, и определяет будущую стоимость акций (я немного утрирую, но в целом это так). Таким образом, добившись высокого объема продаж, стартап готов для проведения IPO по максимально высокой цене — отличный exit (выход из инвестиций) для любого инвестора. Классическая модель, только в ней очень много «если». Если венчурный инвестор вовремя не найдется, или не получится быстро достичь достаточного объема продаж, то пока вы пробиваете железные двери, крупная компания может наладить серийное производство, оставив вас вне игры. Но давайте взглянем на ситуацию по-другому.
Необязательно создавать новый рынок или новую нишу на традиционном — можно встраиваться в ниши, созданные другими. Каждая крупная корпорация создает новые ниши десятками, если не сотнями. Преимущество большой компании не в том, что она инновационна, а в том, что она обеспечена каналами продаж. Я хорошо знаком с бизнесом одной большой компании — знаю, в частности, что она аккумулировала cash в размере 16 млрд долл. Компания занимается высокотехнологичным производством, развивает НИОКР. Так почему бы, казалось бы, не вложить эти деньги в исследования и разработки? Проблема в том, что как только компания заявит о таком вложении, цена на ее акции существенно упадет, ведь НИОКР — вещь рискованная: то ли получится, то ли нет. Таким образом, все крупные корпорации (и это не преувеличение) «сидят» на наличных располагая очень ограниченным набором решений, как этими деньгами распорядиться. Возвращаясь к компании, которую я упомянул в связи с 16 млрд, отмечу, что ее бюджет на собственный НИОКР составляет порядка 2 млрд долл. и примерно в пять раз больше вкладывается в покупку стартапов. И любой крупный высокотехнологичный бизнес тратит на поиск подходящих для покупки активов колоссальное время, силы и средства (Google, без сомнения, станет крупнейшим игроком M&A в 2010 году).
Я уверен, что сотрудничество с большими корпорациями — естественный путь развития российских стартапов. Задача бизнес-инкубаторов и технопарков, как я ее вижу, заключается в том, чтобы корпорации четко сформулировали, в чем заключаются их потребности в области инноваций, и донесли эту информацию до стартапов. А с другой стороны, нужно найти стартапы, способные удовлетворить спрос большого бизнеса, и вырастить их до того состояния, когда они смогут с этой задачей успешно справиться. На глобальном рынке все корпорации заявляют об интересе к инновационным стартапам — на рынке России им еще предстоит этот интерес продекларировать.
Илья Толстов, генеральный директор технопарка «Ингрия»
Читать полностью: http://www.rbcdaily.ru/2010/12/08/focus/562949979263403.shtml
http://www.rbcdaily.ru/2010/12/08/focus/562949979263403
Подписаться на:
Сообщения (Atom)