Intentional stance

From Wikipedia, the free encyclopedia

 

The intentional stance is a term coined by philosopher Daniel Dennett for the level of abstraction in which we view the behavior of an entity in terms of mental properties. It is part of a theory of mental content proposed by Dennett, which provides the underpinnings of his later works on free will, consciousness, folk psychology, and evolution.

Here is how it works: first you decide to treat the object whose behavior is to be predicted as a rational agent; then you figure out what beliefs that agent ought to have, given its place in the world and its purpose. Then you figure out what desires it ought to have, on the same considerations, and finally you predict that this rational agent will act to further its goals in the light of its beliefs. A little practical reasoning from the chosen set of beliefs and desires will in most instances yield a decision about what the agent ought to do; that is what you predict the agent will do.

— Daniel Dennett, The Intentional Stance, p. 17

Dennett and intentionality

Dennett (1971, p. 87) states that he took the concept of "intentionality" from the work of the German philosopher Franz Brentano. When clarifying the distinction between mental phenomena (viz., mental activity) and physical phenomena, Brentano (p. 97) argued that, in contrast with physical phenomena, the "distinguishing characteristic of all mental phenomena" was "the reference to something as an object" – a characteristic he called "intentional inexistence". Dennett constantly speaks of the "aboutness" of intentionality; for example: "the aboutness of the pencil marks composing a shopping list is derived from the intentions of the person whose list it is" (Dennett, 1995, p. 240). 

John Searle (1999, pp. 85) stresses that "competence" in predicting/explaining human behaviour involves being able to both recognize others as "intentional" beings, and interpret others' minds as having "intentional states" (e.g., beliefs and desires):

The primary evolutionary role of the mind is to relate us in certain ways to the environment, and especially to other people. My subjective states relate me to the rest of the world, and the general name of that relationship is "intentionality." These subjective states include beliefs and desires, intentions and perceptions, as well as loves and hates, fears and hopes. "Intentionality," to repeat, is the general term for all the various forms by which the mind can be directed at, or be about, or of, objects and states of affairs in the world. (p.85)

According to Dennett (1987, pp. 48–49), folk psychology provides a systematic, "reason-giving explanation" for a particular action, and an account of the historical origins of that action, based on deeply embedded assumptions about the agent; namely that:

• the agent's action was entirely rational;
• the agent's action was entirely reasonable (in the prevailing circumstances);
• the agent held certain beliefs;
• the agent desired certain things; and
• the agent's future action could be systematically predicted from the beliefs and desires so ascribed.

This approach is also consistent with the earlier work of Fritz Heider and Marianne Simmel, whose joint study revealed that, when subjects were presented with an animated display of 2-dimensional shapes, they were inclined to ascribe intentions to the shapes.

Further, Dennett (1987, p. 52) argues that, based on our fixed personal views of what all humans ought to believe, desire and do, we predict (or explain) the beliefs, desires and actions of others "by calculating in a normative system"; and, driven by the reasonable assumption that all humans are rational beings – who do have specific beliefs and desires and do act on the basis of those beliefs and desires in order to get what they want – these predictions/explanations are based on four simple rules:

1. The agent's beliefs are those a rational individual ought to have (i.e., given their "perceptual capacities", "epistemic needs" and "biography");
2. In general, these beliefs "are both true and relevant to [their] life;
3. The agent's desires are those a rational individual ought to have (i.e., given their "biological needs", and "the most practicable means of satisfying them") in order to further their "survival" and "procreation" needs; and
4. The agent's behaviour will be composed of those acts a rational individual holding those beliefs (and having those desires) ought to perform.

Dennett's three levels

The core idea is that, when understanding, explaining, and/or predicting the behavior of an object, we can choose to view it at varying levels of abstraction. The more concrete the level, the more accurate in principle our predictions are; the more abstract, the greater the computational power we gain by zooming out and skipping over the irrelevant details.

Dennett defines three levels of abstraction, attained by adopting one of three entirely different "stances", or intellectual strategies: the physical stance; the design stance; and the intentional stance:

• The most concrete is the physical stance, the domain of physics and chemistry, which makes predictions from knowledge of the physical constitution of the system and the physical laws that govern its operation; and thus, given a particular set of physical laws and initial conditions, and a particular configuration, a specific future state is predicted (this could also be called the "structure stance"). At this level, we are concerned with such things as mass, energy, velocity, and chemical composition. When we predict where a ball is going to land based on its current trajectory, we are taking the physical stance. Another example of this stance comes when we look at a strip made up of two types of metal bonded together and predict how it will bend as the temperature changes, based on the physical properties of the two metals.
• Somewhat more abstract is the design stance, the domain of biology and engineering, which requires no knowledge of the physical constitution or the physical laws that govern a system's operation. Based on an implicit assumption that there is no malfunction in the system, predictions are made from knowledge of the purpose of the system's design (this could also be called the "teleological stance"). At this level, we are concerned with such things as purpose, function and design. When we predict that a bird will fly when it flaps its wings on the basis that wings are made for flying, we are taking the design stance. Likewise, we can understand the bimetallic strip as a particular type of thermometer, not concerning ourselves with the details of how this type of thermometer happens to work. We can also recognize the purpose that this thermometer serves inside a thermostat and even generalize to other kinds of thermostats that might use a different sort of thermometer. We can even explain the thermostat in terms of what it's good for, saying that it keeps track of the temperature and turns on the heater whenever it gets below a minimum, turning it off once it reaches a maximum.
• Most abstract is the intentional stance, the domain of software and minds, which requires no knowledge of either structure or design, and "[clarifies] the logic of mentalistic explanations of behaviour, their predictive power, and their relation to other forms of explanation" (Bolton & Hill, 1996, p. 24). Predictions are made on the basis of explanations expressed in terms of meaningful mental states; and, given the task of predicting or explaining the behaviour of a specific agent (a person, animal, corporation, artifact, nation, etc.), it is implicitly assumed that the agent will always act on the basis of its beliefs and desires in order to get precisely what it wants (this could also be called the "folk psychology stance"). At this level, we are concerned with such things as belief, thinking and intent. When we predict that the bird will fly away because it knows the cat is coming and is afraid of getting eaten, we are taking the intentional stance. Another example would be when we predict that Mary will leave the theater and drive to the restaurant because she sees that the movie is over and is hungry.
• In 1971, Dennett also postulated that, whilst "the intentional stance presupposes neither lower stance", there may well be a fourth, higher level: a "truly moral stance toward the system" – the "personal stance" – which not only "presupposes the intentional stance" (viz., treats the system as rational) but also "views it as a person" (1971/1978, p. 240).

A key point is that switching to a higher level of abstraction has its risks as well as its benefits. For example, when we view both a bimetallic strip and a tube of mercury as thermometers, we can lose track of the fact that they differ in accuracy and temperature range, leading to false predictions as soon as the thermometer is used outside the circumstances for which it was designed. The actions of a mercury thermometer heated to 500 °C can no longer be predicted on the basis of treating it as a thermometer; we have to sink down to the physical stance to understand it as a melted and boiled piece of junk. For that matter, the "actions" of a dead bird are not predictable in terms of beliefs or desires. 

Even when there is no immediate error, a higher-level stance can simply fail to be useful. If we were to try to understand the thermostat at the level of the intentional stance, ascribing to it beliefs about how hot it is and a desire to keep the temperature just right, we would gain no traction over the problem as compared to staying at the design stance, but we would generate theoretical commitments that expose us to absurdities, such as the possibility of the thermostat not being in the mood to work today because the weather is so nice. Whether to take a particular stance, then, is determined by how successful that stance is when applied. 

Dennett argues that it is best to understand human behavior at the level of the intentional stance, without making any specific commitments to any deeper reality of the artifacts of folk psychology. In addition to the controversy inherent in this, there is also some dispute about the extent to which Dennett is committing to realism about mental properties. Initially, Dennett's interpretation was seen as leaning more towards instrumentalism, but over the years, as this idea has been used to support more extensive theories of consciousness, it has been taken as being more like Realism. His own words hint at something in the middle, as he suggests that the self is as real as a center of gravity, "an abstract object, a theorist's fiction", but operationally valid.

As a way of thinking about things, Dennett's intentional stance is entirely consistent with everyday commonsense understanding; and, thus, it meets Eleanor Rosch's (1978, p. 28) criterion of the "maximum information with the least cognitive effort". Rosch argues that, implicit within any system of categorization, are the assumptions that:

• the major purpose of any system of categorization is to reduce the randomness of the universe by providing "maximum information with the least cognitive effort", and
• the real world is structured and systematic, rather than being arbitrary or unpredictable. Thus, if a particular way of categorizing information does, indeed, "provide maximum information with the least cognitive effort", it can only do so because the structure of that particular system of categories corresponds with the perceived structure of the real world.

Also, the intentional stance meets the criteria Dennett specified (1995, pp. 50–51) for algorithms:

• Substrate Neutrality: It is a "mechanism" that produces results regardless of the material used to perform the procedure ("the power of the procedure is due to its logical structure, not the causal powers of the materials used in the instantiation").
• Underlying Mindlessness: Each constituent step, and each transition between each step, is so utterly simple, that they can be performed by a "dutiful idiot".
• Guaranteed Results: "Whatever it is that an algorithm does, it always does it, if it is executed without misstep. An algorithm is a foolproof recipe."

Variants of Dennett's three stances

The general notion of a three level system was widespread in the late 1970s/early 1980s; for example, when discussing the mental representation of information from a cognitive psychology perspective, Glass and his colleagues (1979, p. 24) distinguished three important aspects of representation:

1. the content ("what is being represented");
2. the code ("the format of the representation"); and
3. the medium ("the physical realization of the code").

Other significant cognitive scientists who also advocated a three level system were Allen Newell, Zenon Pylyshyn, and David Marr. The parallels between the four representations (each of which implicitly assumed that computers and human minds displayed each of the three distinct levels) are detailed in the following table: 

Daniel Dennett "Stances"	Zenon Pylyshyn "Levels of Organization"	Allen Newell "Levels of Description"	David Marr "Levels of Analysis"
Physical Stance.	Physical Level, or Biological Level.	Physical Level, or Device Level.	Hardware Implementation Level.
Design Stance.	Symbol Level.	Program Level, or Symbol Level.	Representation and Algorithm Level.
Intentional Stance.	Semantic, or Knowledge Level.	Knowledge Level.	Computational Theory Level.

Objections and replies

The most obvious objection to Dennett is the intuition that it "matters" to us whether an object has an inner life or not. The claim is that we don't just imagine the intentional states of other people in order to predict their behaviour; the fact that they have thoughts and feelings just like we do is central to notions such as trust, friendship and love. The Blockhead argument proposes that someone, Jones, has a twin who is in fact not a person but a very sophisticated robot which looks and acts like Jones in every way, but who (it is claimed) somehow does not have any thoughts or feelings at all, just a chip which controls his behaviour; in other words, "the lights are on but no one's home". According to the intentional systems theory (IST), Jones and the robot have precisely the same beliefs and desires, but this is claimed to be false. The IST expert assigns the same mental states to Blockhead as he does to Jones, "whereas in fact [Blockhead] has not a thought in his head." Dennett has argued against this by denying the premise, on the basis that the robot is a philosophical zombie and therefore metaphysically impossible. In other words, if something acts in all ways conscious, it necessarily is, as consciousness is defined in terms of behavioral capacity, not ineffable qualia.

Another objection attacks the premise that treating people as ideally rational creatures will yield the best predictions. Stephen Stich argues that people often have beliefs or desires which are irrational or bizarre, and IST doesn't allow us to say anything about these. If the person's "environmental niche" is examined closely enough, and the possibility of malfunction in their brain (which might affect their reasoning capacities) is looked into, it may be possible to formulate a predictive strategy specific to that person. Indeed this is what we often do when someone is behaving unpredictably — we look for the reasons why. In other words, we can only deal with irrationality by contrasting it against the background assumption of rationality. This development significantly undermines the claims of the intentional stance argument. 

The rationale behind the intentional stance is based on evolutionary theory, particularly the notion that the ability to make quick predictions of a system's behaviour based on what we think it might be thinking was an evolutionary adaptive advantage. The fact that our predictive powers are not perfect is a further result of the advantages sometimes accrued by acting contrary to expectations.

Neural evidence

Philip Robbins and Anthony I. Jack suggest that "Dennett's philosophical distinction between the physical and intentional stances has a lot going for it" from the perspective of psychology and neuroscience. They review studies on abilities to adopt an intentional stance (variously called "mindreading," "mentalizing," or "theory of mind") as distinct from adopting a physical stance ("folk physics," "intuitive physics," or "theory of body"). Autism seems to be a deficit in the intentional stance with preservation of the physical stance, while Williams syndrome can involve deficits in the physical stance with preservation of the intentional stance. This tentatively suggests a double dissociation of intentional and physical stances in the brain. However, most studies have found no evidence of impairment in autistic individuals' ability to understand other people's basic intentions or goals; instead, data suggests that impairments are found in understanding more complex social emotions or in considering others' viewpoints.

Robbins and Jack point to a 2003 study in which participants viewed animated geometric shapes in different "vignettes," some of which could be interpreted as constituting social interaction, while others suggested mechanical behavior. Viewing social interactions elicited activity in brain regions associated with identifying faces and biological objects (posterior temporal cortex), as well as emotion processing (right amygdala and ventromedial prefrontal cortex). Meanwhile, the mechanical interactions activated regions related to identifying objects like tools that can be manipulated (posterior temporal lobe). The authors suggest "that these findings reveal putative 'core systems' for social and mechanical understanding that are divisible into constituent parts or elements with distinct processing and storage capabilities."

Phenomenal stance

Robbins and Jack argue for an additional stance beyond the three that Dennett outlined. They call it the phenomenal stance: Attributing consciousness, emotions, and inner experience to a mind. The explanatory gap of the hard problem of consciousness illustrates this tendency of people to see phenomenal experience as different from physical processes. The authors suggest that psychopathy may represent a deficit in the phenomenal but not intentional stance, while people with autism appear to have intact moral sensibilities, just not mind-reading abilities. These examples suggest a double dissociation between the intentional and phenomenal stances.

In a follow-up paper, Robbins and Jack describe four experiments about how the intentional and phenomenal stances relate to feelings of moral concern. The first two experiments showed that talking about lobsters as strongly emotional led to a much greater sentiment that lobsters deserved welfare protections than did talking about lobsters as highly intelligent. The third and fourth studies found that perceiving an agent as vulnerable led to greater attributions of phenomenal experience. Also, people who scored higher on the empathetic-concern subscale of the Interpersonal Reactivity Index had generally higher absolute attributions of mental experience.

Bryce Huebner (2010) performed two experimental philosophy studies to test students' ascriptions of various mental states to humans compared with cyborgs and robots. Experiment 1 showed that while students attributed both beliefs and pains most strongly to humans, they were more willing to attribute beliefs than pains to robots and cyborgs. "[T]hese data seem to confirm that commonsense psychology does draw a distinction between phenomenal and non-phenomenal states--and this distinction seems to be dependent on the structural properties of an entity in a way that ascriptions of non-phenomenal states are not." However, this conclusion is only tentative in view of the high variance among participants. Experiment 2 showed analogous results: Both beliefs and happiness were ascribed most strongly to biological humans, and ascriptions of happiness to robots or cyborgs were less common than ascriptions of beliefs.

Darwin's Dangerous Idea (updated as The Library of Mendel)

From Wikipedia, the free encyclopedia

Darwin's Dangerous Idea

Cover of the first edition
Author	Daniel C. Dennett
Country	United States
Language	English
Subjects	Evolution Ethics
Published	1995 (Simon & Schuster)
Media type	Print (Hardcover and Paperback)
Pages	586

Darwin's Dangerous Idea: Evolution and the Meanings of Life is a 1995 book by Daniel Dennett, in which the author looks at some of the repercussions of Darwinian theory. The crux of the argument is that, whether or not Darwin's theories are overturned, there is no going back from the dangerous idea that design (purpose or what something is for) might not need a designer. Dennett makes this case on the basis that natural selection is a blind process, which is nevertheless sufficiently powerful to explain the evolution of life. Darwin's discovery was that the generation of life worked algorithmically, that processes behind it work in such a way that given these processes the results that they tend toward must be so.

Dennett says, for example, that by claiming that minds cannot be reduced to purely algorithmic processes, many of his eminent contemporaries are claiming that miracles can occur. These assertions have generated a great deal of debate and discussion in the general public. The book was a finalist for the 1995 National Book Award in non-fiction and the 1996 Pulitzer Prize for General Non-Fiction.

Background

Dennett's previous book was Consciousness Explained (1991). Dennett noted discomfort with Darwinism among not only lay people but also even academics and decided it was time to write a book dealing with the subject. Darwin's Dangerous Idea is not meant to be a work of science, but rather an interdisciplinary book; Dennett admits that he does not understand all of the scientific details himself. He goes into a moderate level of detail, but leaves it for the reader to go into greater depth if desired, providing references to this end.

In writing the book, Dennett wanted to "get thinkers in other disciplines to take evolutionary theory seriously, to show them how they have been underestimating it, and to show them why they have been listening to the wrong sirens". To do this he tells a story; one that is mainly original but includes some material from his previous work.

Dennett taught an undergraduate seminar at Tufts University on Darwin and philosophy, which included most of the ideas in the book. He also had the help of fellow staff and other academics, some of whom read drafts of the book. It is dedicated to W. V. O. Quine, "teacher and friend".

Synopsis

Part I: Starting in the Middle

"Starting in the Middle", Part I of Darwin's Dangerous Idea, gets its name from a quote by Willard Van Orman Quine: "Analyze theory-building how we will, we all must start in the middle. Our conceptual firsts are middle-sized, middle-distance objects, and our introduction to them and to everything comes midway in the cultural evolution of the race." 

The first chapter "Tell Me Why" is named after a song.

Tell me why the stars do shine,
Tell me why the ivy twines,
Tell me why the sky's so blue.
Then I will tell you just why I love you.
Because God made the stars to shine,
Because God made the ivy twine,
Because God made the sky so blue.
Because God made you, that's why I love you.

Before Charles Darwin, and still today, a majority of people see God as the ultimate cause of all design, or the ultimate answer to 'why?' questions. John Locke argued for the primacy of mind before matter, and David Hume, while exposing problems with Locke's view, could not see any alternative.

Darwin's Dangerous Idea makes extensive use of cranes as an analogy.

Darwin provided just such an alternative: evolution. Besides providing evidence of common descent, he introduced a mechanism to explain it: natural selection. According to Dennett, natural selection is a mindless, mechanical and algorithmic process—Darwin's dangerous idea. The third chapter introduces the concept of "skyhooks" and "cranes" (see below). He suggests that resistance to Darwinism is based on a desire for skyhooks, which do not really exist. According to Dennett, good reductionists explain apparent design without skyhooks; greedy reductionists try to explain it without cranes.

Chapter 4 looks at the tree of life, such as how it can be visualized and some crucial events in life's history. The next chapter concerns the possible and the actual, using the 'Library of Mendel' (the space of all logically possible genomes) as a conceptual aid.

In the last chapter of part I, Dennett treats human artifacts and culture as a branch of a unified Design Space. Descent or homology can be detected by shared design features that would be unlikely to appear independently. However, there are also "Forced Moves" or "Good Tricks" that will be discovered repeatedly, either by natural selection or human investigation.

Part II: Darwinian Thinking in Biology

Tree diagram in Origin

The first chapter of part II, "Darwinian Thinking in Biology", asserts that life originated without any skyhooks, and the orderly world we know is the result of a blind and undirected shuffle through chaos.

The eighth chapter's message is conveyed by its title, "Biology is Engineering"; biology is the study of design, function, construction and operation. However, there are some important differences between biology and engineering. Related to the engineering concept of optimization, the next chapter deals with adaptationism, which Dennett endorses, calling Gould and Lewontin's "refutation" of it an illusion. Dennett thinks adaptationism is, in fact, the best way of uncovering constraints.

The tenth chapter, entitled "Bully for Brontosaurus", is an extended critique of Stephen Jay Gould, who Dennett feels has created a distorted view of evolution with his popular writings; his "self-styled revolutions" against adaptationism, gradualism and other orthodox Darwinism all being false alarms. The final chapter of part II dismisses directed mutation, the inheritance of acquired traits and Teilhard's "Omega Point", and insists that other controversies and hypotheses (like the unit of selection and Panspermia) have no dire consequences for orthodox Darwinism.

Part III: Mind, Meaning, Mathematics and Morality

The frontispiece to Thomas Hobbes' Leviathan, which appears at the beginning of chapter 16 "On the Origin of Morality".

"Mind, Meaning, Mathematics and Morality" is the name of Part III, which begins with a quote from Nietzsche. Chapter 12, "The Cranes of Culture", discusses cultural evolution. It asserts that the meme has a role to play in our understanding of culture, and that it allows humans, alone among animals, to "transcend" our selfish genes. "Losing Our Minds to Darwin" follows, a chapter about the evolution of brains, minds and language. Dennett criticizes Noam Chomsky's perceived resistance to the evolution of language, its modeling by artificial intelligence, and reverse engineering.

The evolution of meaning is then discussed, and Dennett uses a series of thought experiments to persuade the reader that meaning is the product of meaningless, algorithmic processes.

Von Kempelen's chess automaton, discussed in chapter 15.

Chapter 15 asserts that Gödel's Theorem does not make certain sorts of artificial intelligence impossible. Dennett extends his criticism to Roger Penrose. The subject then moves on to the origin and evolution of morality, beginning with Thomas Hobbes (who Dennett calls "the first sociobiologist") and Friedrich Nietzsche. He concludes that only an evolutionary analysis of ethics makes sense, though he cautions against some varieties of 'greedy ethical reductionism'. Before moving to the next chapter, he discusses some sociobiology controversies.

The penultimate chapter, entitled "Redesigning Morality", begins by asking if ethics can be 'naturalized'. Dennett does not believe there is much hope of discovering an algorithm for doing the right thing, but expresses optimism in our ability to design and redesign our approach to moral problems. In "The Future of an Idea", the book's last chapter, Dennett praises biodiversity, including cultural diversity. In closing, he uses Beauty and the Beast as an analogy; although Darwin's idea may seem dangerous, it is actually quite beautiful.

Central concepts

Design Space

Dennett believes there is little or no principled difference between the naturally generated products of evolution and the man-made artifacts of human creativity and culture. For this reason he indicates deliberately that the complex fruits of the tree of life are in a very meaningful sense "designed"—even though he does not believe evolution was guided by a higher intelligence. 

Dennett supports using the notion of memes to better understand cultural evolution. He also believes even human creativity might operate by the Darwinian mechanism. This leads him to propose that the "space" describing biological "design" is connected with the space describing human culture and technology.

A precise mathematical definition of Design Space is not given in Darwin's Dangerous Idea. Dennett acknowledges this and admits he is offering a philosophical idea rather than a scientific formulation.

Natural selection as an algorithm

Dennett describes natural selection as a substrate-neutral, mindless algorithm for moving through Design Space.

Universal acid

Dennett writes about the fantasy of a "universal acid" as a liquid that is so corrosive that it would eat through anything that it came into contact with, even a potential container. Such a powerful substance would transform everything it was applied to; leaving something very different in its wake. This is where Dennett draws parallels from the “universal acid” to Darwin’s idea:

...it eats through just about every traditional concept, and leaves in its wake a revolutionized world-view, with most of the old landmarks still recognizable, but transformed in fundamental ways.

While there are people who would like to see Darwin’s idea contained within the field of biology, Dennett asserts that this dangerous idea inevitably “leaks” out to transform other fields as well.

Skyhooks and cranes

Dennett uses the term "skyhook" to describe a source of design complexity that does not build on lower, simpler layers—in simple terms, a miracle. 

In philosophical arguments concerning the reducibility (or otherwise) of the human mind, Dennett's concept pokes fun at the idea of intelligent design emanating from on high, either originating from one or more gods, or providing its own grounds in an absurd, Munchausen-like bootstrapping manner.

Dennett also accuses various competing neo-Darwinian ideas of making use of such supposedly unscientific skyhooks in explaining evolution, coming down particularly hard on the ideas of Stephen Jay Gould.

Dennett contrasts theories of complexity that require such miracles with those based on "cranes", structures that permit the construction of entities of greater complexity but are themselves founded solidly "on the ground" of physical science.

Reception

In The New York Review of Books, John Maynard Smith praised Darwin's Dangerous Idea:

It is therefore a pleasure to meet a philosopher who understands what Darwinism is about, and approves of it. Dennett goes well beyond biology. He sees Darwinism as a corrosive acid, capable of dissolving our earlier belief and forcing a reconsideration of much of sociology and philosophy. Although modestly written, this is not a modest book. Dennett argues that, if we understand Darwin's dangerous idea, we are forced to reject or modify much of our current intellectual baggage...

Writing in the same publication, Stephen Jay Gould criticised Darwin's Dangerous Idea for being an "influential but misguided ultra-Darwinian manifesto":

Daniel Dennett devotes the longest chapter in Darwin's Dangerous Idea to an excoriating caricature of my ideas, all in order to bolster his defense of Darwinian fundamentalism. If an argued case can be discerned at all amid the slurs and sneers, it would have to be described as an effort to claim that I have, thanks to some literary skill, tried to raise a few piddling, insignificant, and basically conventional ideas to "revolutionary" status, challenging what he takes to be the true Darwinian scripture. Since Dennett shows so little understanding of evolutionary theory beyond natural selection, his critique of my work amounts to little more than sniping at false targets of his own construction. He never deals with my ideas as such, but proceeds by hint, innuendo, false attribution, and error.

Gould was also a harsh critic of Dennett's idea of the "universal acid" of natural selection and of his subscription to the idea of memetics; Dennett responded, and the exchange between Dennett, Gould, and Robert Wright was printed in the New York Review of Books.

Biologist H. Allen Orr wrote a critical review emphasizing similar points in the Boston Review.

The book has also provoked a negative reaction from creationists; Frederick Crews writes that Darwin's Dangerous Idea "rivals Richard Dawkins's The Blind Watchmaker as the creationists' most cordially hated text."

The Library of Babel

From Wikipedia, the free encyclopedia

"The Library of Babel"
English language cover
Author	Jorge Luis Borges
Original title	"La biblioteca de Babel"
Translator	numerous
Country	Argentina
Language	Spanish
Genre(s)	Fantasy
Published in	El Jardín de senderos que se bifurcan
Publisher	Editorial Sur
Publication date	1941
Published in English	1962

"The Library of Babel" (Spanish: La biblioteca de Babel) is a short story by Argentine author and librarian Jorge Luis Borges (1899–1986), conceiving of a universe in the form of a vast library containing all possible 410-page books of a certain format and character set.

The story was originally published in Spanish in Borges' 1941 collection of stories El Jardín de senderos que se bifurcan (The Garden of Forking Paths). That entire book was, in turn, included within his much-reprinted Ficciones (1944). Two English-language translations appeared approximately simultaneously in 1962, one by James E. Irby in a diverse collection of Borges's works titled Labyrinths and the other by Anthony Kerrigan as part of a collaborative translation of the entirety of Ficciones.

Plot summary

Borges' narrator describes how his universe consists of an enormous expanse of adjacent hexagonal rooms, each of which contains the bare necessities for human survival—and four walls of bookshelves. Though the order and content of the books are random and apparently completely meaningless, the inhabitants believe that the books contain every possible ordering of just 25 basic characters (22 letters, the period, the comma, and space). Though the vast majority of the books in this universe are pure gibberish, the library also must contain, somewhere, every coherent book ever written, or that might ever be written, and every possible permutation or slightly erroneous version of every one of those books. The narrator notes that the library must contain all useful information, including predictions of the future, biographies of any person, and translations of every book in all languages. Conversely, for many of the texts, some language could be devised that would make it readable with any of a vast number of different contents.

Despite—indeed, because of—this glut of information, all books are totally useless to the reader, leaving the librarians in a state of suicidal despair. This leads some librarians to superstitious and cult-like behaviors, such as the "Purifiers", who arbitrarily destroy books they deem nonsense as they scour through the library seeking the "Crimson Hexagon" and its illustrated, magical books. Others believe that since all books exist in the library, somewhere one of the books must be a perfect index of the library's contents; some even believe that a messianic figure known as the "Man of the Book" has read it, and they travel through the library seeking him.

Themes

Borges in 1976

The story repeats the theme of Borges' 1939 essay "The Total Library" ("La Biblioteca total"), which in turn acknowledges the earlier development of this theme by Kurd Lasswitz in his 1901 story "The Universal Library" ("Die Universalbibliothek"):

Certain examples that Aristotle attributes to Democritus and Leucippus clearly prefigure it, but its belated inventor is Gustav Theodor Fechner, and its first exponent, Kurd Lasswitz. [...] In his book The Race with the Tortoise (Berlin, 1919), Dr Theodor Wolff suggests that it is a derivation from, or a parody of, Ramón Llull's thinking machine [...T]he elements of his game are the universal orthographic symbols, not the words of a language [...] Lasswitz arrives at twenty-five symbols (twenty-two letters, the space, the period, the comma), whose recombinations and repetitions encompass everything possible to express in all languages. The totality of such variations would form a Total Library of astronomical size. Lasswitz urges mankind to construct that inhuman library, which chance would organize and which would eliminate intelligence. (Wolff's The Race with the Tortoise expounds the execution and the dimensions of that impossible enterprise.)

Many of Borges' signature motifs are featured in the story, including infinity, reality, cabalistic reasoning, and labyrinths. The concept of the library is often compared to Borel's dactylographic monkey theorem. There is no reference to monkeys or typewriters in "The Library of Babel", although Borges had mentioned that analogy in "The Total Library": "[A] half-dozen monkeys provided with typewriters would, in a few eternities, produce all the books in the British Museum." In this story, the closest equivalent is the line, "A blasphemous sect suggested [...] that all men should juggle letters and symbols until they constructed, by an improbable gift of chance, these canonical books."

Borges would examine a similar idea in his 1975 story, "The Book of Sand" in which there is an infinite book (or book with an indefinite number of pages) rather than an infinite library. Moreover, the story's Book of Sand is said to be written in an unknown alphabet and its content is not obviously random. In The Library of Babel, Borges interpolates Italian mathematician Bonaventura Cavalieri's suggestion that any solid body could be conceptualized as the superimposition of an infinite number of planes.

The concept of the library is also overtly analogous to the view of the universe as a sphere having its center everywhere and its circumference nowhere. The mathematician and philosopher Blaise Pascal employed this metaphor, and in an earlier essay Borges noted that Pascal's manuscript called the sphere effroyable, or "frightful".

In any case, a library containing all possible books, arranged at random, might as well be a library containing zero books, as any true information would be buried in, and rendered indistinguishable from, all possible forms of false information; the experience of opening to any page of any of the library's books has been simulated by websites which create screenfuls of random letters.

The quote at the beginning of the story, "By this art you may contemplate the variation of the twenty-three letters," is from Robert Burton's 1621 The Anatomy of Melancholy.

Philosophical implications

There are numerous philosophical implications within the idea of the finite library which exhausts all possibilities. Every book in the library is "intelligible" if one decodes it correctly, simply because it can be decoded from any other book in the library using a third book as a one-time pad. This lends itself to the philosophical idea proposed by Immanuel Kant, that our mind helps to structure our experience of reality; thus the rules of reality (as we know it) are intrinsic to the mind. So if we identify these rules, we can better decode 'reality'. One might speculate that these rules are contained in the crimson hexagon room which is the key to decoding the others. The library becomes a temptation, even an obsession, because it contains these gems of enlightenment while also burying them in deception. On a psychological level, the infinite storehouse of information is a hindrance and a distraction, because it lures one away from writing one's own book (i.e. living one's life). Anything one might write would of course already exist. One can see any text as being pulled from the library by the act of the author defining the search letter by letter until they reach a text close enough to the one they intended to write. The text already existed theoretically, but had to be found by the act of the author's imagination. Another implication is an argument against certain proofs of the existence of God, as it is carried out by David Hume using the thought experiment of a similar library of books generated not by human mind, but by nature.

Infinite extent

In mainstream theories of natural language syntax, every syntactically-valid utterance can be extended to produce a new, longer one, because of recursion. If this process can be continued indefinitely, then there is no upper bound on the length of a well-formed utterance and the number of unique well-formed strings of any language is countably infinite. However, the books in the Library of Babel are of bounded length ("each book is of four hundred and ten pages; each page, of forty lines, each line, of some eighty letters"), so the Library can only contain a finite number of distinct strings, and thus cannot contain all possible well-formed utterances. Borges' narrator notes this fact, but believes that the Library is nevertheless infinite; he speculates that it repeats itself periodically, given an eventual "order" to the "disorder" of the seemingly-random arrangement of books.

Quine's reduction

In a short essay, W. V. O. Quine noted the interesting fact that the Library of Babel is finite (that is, we will theoretically come to a point in history where everything has been written), and that the Library of Babel can be constructed in its entirety simply by writing a dot on one piece of paper and a dash on another. These two sheets of paper could then be alternated at random to produce every possible text, in Morse code or equivalently binary. Writes Quine, "The ultimate absurdity is now staring us in the face: a universal library of two volumes, one containing a single dot and the other a dash. Persistent repetition and alternation of the two are sufficient, we well know, for spelling out any and every truth. The miracle of the finite but universal library is a mere inflation of the miracle of binary notation: everything worth saying, and everything else as well, can be said with two characters."

Comparison with biology

The full possible set of protein sequences (Protein sequence space) has been compared to the Library of Babel. In the Library of Babel, finding any book that made sense was impossible due to the sheer number and lack of order. The same would be true of protein sequences if it were not for natural selection, which has picked out only protein sequences that make sense. Additionally, each protein sequence is surrounded by a set of neighbors (point mutants) that are likely to have at least some function. Daniel Dennett's 1995 book Darwin's Dangerous Idea includes an elaboration of the Library of Babel concept to imagine the set of all possible genetic sequences, which he calls the Library of Mendel, in order to illustrate the mathematics of genetic variation. Dennett uses this concept again later in the book to imagine all possible algorithms that can be included in his Toshiba computer, which he calls the Library of Toshiba. He describes the Library of Mendel and the Library of Toshiba as subsets within the Library of Babel.

Influence on later writers

• Umberto Eco's postmodern novel The Name of the Rose (1980) features a labyrinthine library, presided over by a blind monk named Jorge of Burgos.
• In "The Net of Babel", published in Interzone in 1995, David Langford imagines the Library becoming computerized for easy access. This aids the librarians in searching for specific text while also highlighting the futility of such searches as they can find anything, but nothing of meaning as such. The sequel continues many of Borges's themes, while also highlighting the difference between data and information, and satirizing the Internet.
• Russell Standish's Theory of Nothing uses the concept of the Library of Babel to illustrate how an ultimate ensemble containing all possible descriptions would in sum contain zero information and would thus be the simplest possible explanation for the existence of the universe. This theory, therefore, implies the reality of all universes.
• Michael Ende reused the idea of a universe of hexagonal rooms in the Temple of a Thousand Doors from The Neverending Story, which contained all the possible characteristics of doors in the fantastic realm. A later chapter features the infinite monkey theorem.
• Terry Pratchett uses the concept of the infinite library in his Discworld novels. The knowledgeable librarian is a human wizard transformed into an orangutan.
• The Unimaginable Mathematics of Borges' Library of Babel (2008) by William Goldbloom Bloch explores the short story from a mathematical perspective. Bloch analyzes the hypothetical library presented by Borges using the ideas of topology, information theory, and geometry.
• In Greg Bear's novel City at the End of Time (2008), the sum-runners carried by the protagonists are intended by their creator to be combined to form a 'Babel', an infinite library containing every possible permutation of every possible character in every possible language. Bear has stated that this was inspired by Borges, who is also namechecked in the novel. Borges is described as an unknown Argentinian who commissioned an encyclopedia of impossible things, a reference to either "Tlön, Uqbar, Orbis Tertius" or the Book of Imaginary Beings.
• Fone, a short comic novel drawn by Milo Manara, features a human astronaut and his alien partner stranded on a planet named Borges Profeta. The planet is overflowed by books containing all the possible permutations of letters.
• Steven L. Peck wrote a novella entitled A Short Stay in Hell (2012) in which the protagonist must find the book containing his life story in an afterlife replica of Borges' Library of Babel.
• The third season of Carmilla, a Canadian single-frame web series based on the novella by J. Sheridan Le Fanu, is set in a mystical library described as "non-Euclidean" and omnipotent. It contains a door that, depending on the knocking pattern on its panels, can be opened into any universe. It also creates a temporary parallel universe and is able to shift a character between the parallel and the original. As the parallel universe collapses, darkness falls, and a character perishes in the void after uttering the words, "O time thy pyramids," which are contained on the second-to-last page of a book in the Library of Babel.
• In Christopher Nolan's film Interstellar, the protagonist, Cooper, played by Matthew McConaughey, becomes trapped in a world which mirrors that of Borges' i.e. Cooper's universe consists of an enormous expanse of adjacent hexagonal rooms, or libraries, each of which contains the bare necessities for survival. Though the order and content of the books and rooms are random and apparently completely meaningless, Cooper can, by manipulating the books, affect change in the "real" world and is, as such, analogous to the "Man of the Book", the messianic figure in The Library of Babel. Unlike the Man of the Book, however, Cooper is something more than just a metaphor and has a transformative role in his Universe, becoming a catalyst and an agent of change.
• Jonathan Basile enterprised to recreate the Library in Borges' story on his website http://libraryofbabel.info, adapted to the English language. An algorithm he created generates a 'book' by iterating every permutation of 29 characters: the 26 English letters, space, comma, and period. Each book is marked by a coordinate, corresponding to its place on the hexagonal library (hexagon name, wall number, shelf number, and book name) so that every book can be found at the same place every time. The website is said to contain "all possible pages of 3200 characters, about 10⁴⁶⁷⁷ books".

Design of experiments

From Wikipedia, the free encyclopedia

Design of experiments with full factorial design (left), response surface with second-degree polynomial (right)

The design of experiments (DOE, DOX, or experimental design) is the design of any task that aims to describe or explain the variation of information under conditions that are hypothesized to reflect the variation. The term is generally associated with experiments in which the design introduces conditions that directly affect the variation, but may also refer to the design of quasi-experiments, in which natural conditions that influence the variation are selected for observation.

In its simplest form, an experiment aims at predicting the outcome by introducing a change of the preconditions, which is represented by one or more independent variables, also referred to as "input variables" or "predictor variables." The change in one or more independent variables is generally hypothesized to result in a change in one or more dependent variables, also referred to as "output variables" or "response variables." The experimental design may also identify control variables that must be held constant to prevent external factors from affecting the results. Experimental design involves not only the selection of suitable independent, dependent, and control variables, but planning the delivery of the experiment under statistically optimal conditions given the constraints of available resources. There are multiple approaches for determining the set of design points (unique combinations of the settings of the independent variables) to be used in the experiment.

Main concerns in experimental design include the establishment of validity, reliability, and replicability. For example, these concerns can be partially addressed by carefully choosing the independent variable, reducing the risk of measurement error, and ensuring that the documentation of the method is sufficiently detailed. Related concerns include achieving appropriate levels of statistical power and sensitivity.

Correctly designed experiments advance knowledge in the natural and social sciences and engineering. Other applications include marketing and policy making.

History

Systematic clinical trials

In 1747, while serving as surgeon on HMS Salisbury, James Lind carried out a systematic clinical trial to compare remedies for scurvy. This systematic clinical trial constitutes a type of DOE.

Lind selected 12 men from the ship, all suffering from scurvy. Lind limited his subjects to men who "were as similar as I could have them," that is, he provided strict entry requirements to reduce extraneous variation. He divided them into six pairs, giving each pair different supplements to their basic diet for two weeks. The treatments were all remedies that had been proposed:

• A quart of cider every day.
• Twenty five gutts (drops) of vitriol (sulphuric acid) three times a day upon an empty stomach.
• One half-pint of seawater every day.
• A mixture of garlic, mustard, and horseradish in a lump the size of a nutmeg.
• Two spoonfuls of vinegar three times a day.
• Two oranges and one lemon every day.

The citrus treatment stopped after six days when they ran out of fruit, but by that time one sailor was fit for duty while the other had almost recovered. Apart from that, only group one (cider) showed some effect of its treatment. The remainder of the crew presumably served as a control, but Lind did not report results from any control (untreated) group.

Statistical experiments, following Charles S. Peirce

A theory of statistical inference was developed by Charles S. Peirce in "Illustrations of the Logic of Science" (1877–1878) and "A Theory of Probable Inference" (1883), two publications that emphasized the importance of randomization-based inference in statistics.

Randomized experiments

Charles S. Peirce randomly assigned volunteers to a blinded, repeated-measures design to evaluate their ability to discriminate weights. Peirce's experiment inspired other researchers in psychology and education, which developed a research tradition of randomized experiments in laboratories and specialized textbooks in the 1800s.

Optimal designs for regression models

Charles S. Peirce also contributed the first English-language publication on an optimal design for regression models in 1876. A pioneering optimal design for polynomial regression was suggested by Gergonne in 1815. In 1918, Kirstine Smith published optimal designs for polynomials of degree six (and less).

Sequences of experiments

The use of a sequence of experiments, where the design of each may depend on the results of previous experiments, including the possible decision to stop experimenting, is within the scope of Sequential analysis, a field that was pioneered by Abraham Wald in the context of sequential tests of statistical hypotheses. Herman Chernoff wrote an overview of optimal sequential designs, while adaptive designs have been surveyed by S. Zacks. One specific type of sequential design is the "two-armed bandit", generalized to the multi-armed bandit, on which early work was done by Herbert Robbins in 1952.

Fisher's principles

A methodology for designing experiments was proposed by Ronald Fisher, in his innovative books: The Arrangement of Field Experiments (1926) and The Design of Experiments (1935). Much of his pioneering work dealt with agricultural applications of statistical methods. As a mundane example, he described how to test the lady tasting tea hypothesis, that a certain lady could distinguish by flavour alone whether the milk or the tea was first placed in the cup. These methods have been broadly adapted in the physical and social sciences, are still used in agricultural engineering and differ from the design and analysis of computer experiments.

Comparison

In some fields of study it is not possible to have independent measurements to a traceable metrology standard. Comparisons between treatments are much more valuable and are usually preferable, and often compared against a scientific control or traditional treatment that acts as baseline.

Randomization

Random assignment is the process of assigning individuals at random to groups or to different groups in an experiment, so that each individual of the population has the same chance of becoming a participant in the study. The random assignment of individuals to groups (or conditions within a group) distinguishes a rigorous, "true" experiment from an observational study or "quasi-experiment". There is an extensive body of mathematical theory that explores the consequences of making the allocation of units to treatments by means of some random mechanism (such as tables of random numbers, or the use of randomization devices such as playing cards or dice). Assigning units to treatments at random tends to mitigate confounding, which makes effects due to factors other than the treatment to appear to result from the treatment.

The risks associated with random allocation (such as having a serious imbalance in a key characteristic between a treatment group and a control group) are calculable and hence can be managed down to an acceptable level by using enough experimental units. However, if the population is divided into several subpopulations that somehow differ, and the research requires each subpopulation to be equal in size, stratified sampling can be used. In that way, the units in each subpopulation are randomized, but not the whole sample. The results of an experiment can be generalized reliably from the experimental units to a larger statistical population of units only if the experimental units are a random sample from the larger population; the probable error of such an extrapolation depends on the sample size, among other things.

Statistical replication

Measurements are usually subject to variation and measurement uncertainty; thus they are repeated and full experiments are replicated to help identify the sources of variation, to better estimate the true effects of treatments, to further strengthen the experiment's reliability and validity, and to add to the existing knowledge of the topic. However, certain conditions must be met before the replication of the experiment is commenced: the original research question has been published in a peer-reviewed journal or widely cited, the researcher is independent of the original experiment, the researcher must first try to replicate the original findings using the original data, and the write-up should state that the study conducted is a replication study that tried to follow the original study as strictly as possible.

Blocking

Blocking is the non-random arrangement of experimental units into groups (blocks/lots) consisting of units that are similar to one another. Blocking reduces known but irrelevant sources of variation between units and thus allows greater precision in the estimation of the source of variation under study.

Orthogonality

Example of orthogonal factorial design

Orthogonality concerns the forms of comparison (contrasts) that can be legitimately and efficiently carried out. Contrasts can be represented by vectors and sets of orthogonal contrasts are uncorrelated and independently distributed if the data are normal. Because of this independence, each orthogonal treatment provides different information to the others. If there are T treatments and T – 1 orthogonal contrasts, all the information that can be captured from the experiment is obtainable from the set of contrasts.

Factorial experiments

Use of factorial experiments instead of the one-factor-at-a-time method. These are efficient at evaluating the effects and possible interactions of several factors (independent variables). Analysis of experiment design is built on the foundation of the analysis of variance, a collection of models that partition the observed variance into components, according to what factors the experiment must estimate or test.

Example

This example is attributed to Harold Hotelling. It conveys some of the flavor of those aspects of the subject that involve combinatorial designs.

Weights of eight objects are measured using a pan balance and set of standard weights. Each weighing measures the weight difference between objects in the left pan vs. any objects in the right pan by adding calibrated weights to the lighter pan until the balance is in equilibrium. Each measurement has a random error. The average error is zero; the standard deviations of the probability distribution of the errors is the same number σ on different weighings; errors on different weighings are independent. Denote the true weights by

${\displaystyle \theta _{1},\dots ,\theta _{8}.\,}$

We consider two different experiments:

1. Weigh each object in one pan, with the other pan empty. Let X_i be the measured weight of the object, for i = 1, ..., 8.
2. Do the eight weighings according to the following schedule and let Y_i be the measured difference for i = 1, ..., 8:

${\displaystyle {\begin{array}{lcc}&{\text{left pan}}&{\text{right pan}}\\\hline {\text{1st weighing:}}&1\ 2\ 3\ 4\ 5\ 6\ 7\ 8&{\text{(empty)}}\\{\text{2nd:}}&1\ 2\ 3\ 8\ &4\ 5\ 6\ 7\\{\text{3rd:}}&1\ 4\ 5\ 8\ &2\ 3\ 6\ 7\\{\text{4th:}}&1\ 6\ 7\ 8\ &2\ 3\ 4\ 5\\{\text{5th:}}&2\ 4\ 6\ 8\ &1\ 3\ 5\ 7\\{\text{6th:}}&2\ 5\ 7\ 8\ &1\ 3\ 4\ 6\\{\text{7th:}}&3\ 4\ 7\ 8\ &1\ 2\ 5\ 6\\{\text{8th:}}&3\ 5\ 6\ 8\ &1\ 2\ 4\ 7\end{array}}}$

Then the estimated value of the weight θ₁ is

${\displaystyle {\widehat {\theta }}_{1}={\frac {Y_{1}+Y_{2}+Y_{3}+Y_{4}-Y_{5}-Y_{6}-Y_{7}-Y_{8}}{8}}.}$

Similar estimates can be found for the weights of the other items. For example

${\displaystyle {\begin{aligned}{\widehat {\theta }}_{2}&={\frac {Y_{1}+Y_{2}-Y_{3}-Y_{4}+Y_{5}+Y_{6}-Y_{7}-Y_{8}}{8}}.\\[5pt]{\widehat {\theta }}_{3}&={\frac {Y_{1}+Y_{2}-Y_{3}-Y_{4}-Y_{5}-Y_{6}+Y_{7}+Y_{8}}{8}}.\\[5pt]{\widehat {\theta }}_{4}&={\frac {Y_{1}-Y_{2}+Y_{3}-Y_{4}+Y_{5}-Y_{6}+Y_{7}-Y_{8}}{8}}.\\[5pt]{\widehat {\theta }}_{5}&={\frac {Y_{1}-Y_{2}+Y_{3}-Y_{4}-Y_{5}+Y_{6}-Y_{7}+Y_{8}}{8}}.\\[5pt]{\widehat {\theta }}_{6}&={\frac {Y_{1}-Y_{2}-Y_{3}+Y_{4}+Y_{5}-Y_{6}-Y_{7}+Y_{8}}{8}}.\\[5pt]{\widehat {\theta }}_{7}&={\frac {Y_{1}-Y_{2}-Y_{3}+Y_{4}-Y_{5}+Y_{6}+Y_{7}-Y_{8}}{8}}.\\[5pt]{\widehat {\theta }}_{8}&={\frac {Y_{1}+Y_{2}+Y_{3}+Y_{4}+Y_{5}+Y_{6}+Y_{7}+Y_{8}}{8}}.\end{aligned}}}$

The question of design of experiments is: which experiment is better?

The variance of the estimate X₁ of θ₁ is σ² if we use the first experiment. But if we use the second experiment, the variance of the estimate given above is σ²/8. Thus the second experiment gives us 8 times as much precision for the estimate of a single item, and estimates all items simultaneously, with the same precision. What the second experiment achieves with eight would require 64 weighings if the items are weighed separately. However, note that the estimates for the items obtained in the second experiment have errors that correlate with each other.

Many problems of the design of experiments involve combinatorial designs, as in this example and others.

Avoiding false positives

False positive conclusions, often resulting from the pressure to publish or the author's own confirmation bias, are an inherent hazard in many fields. A good way to prevent biases potentially leading to false positives in the data collection phase is to use a double-blind design. When a double-blind design is used, participants are randomly assigned to experimental groups but the researcher is unaware of what participants belong to which group. Therefore, the researcher can not affect the participants' response to the intervention. Experimental designs with undisclosed degrees of freedom are a problem. This can lead to conscious or unconscious "p-hacking": trying multiple things until you get the desired result. It typically involves the manipulation - perhaps unconsciously - of the process of statistical analysis and the degrees of freedom until they return a figure below the p less than 0.05 level of statistical significance. So the design of the experiment should include a clear statement proposing the analyses to be undertaken. P-hacking can be prevented by preregistering researches, in which researchers have to send their data analysis plan to the journal they wish to publish their paper in before they even start their data collection, so no data manipulation is possible (https://osf.io). Another way to prevent this is taking the double-blind design to the data-analysis phase, where the data are sent to a data-analyst unrelated to the research who scrambles up the data so there is no way to know which participants belong to before they are potentially taken away as outliers.

Clear and complete documentation of the experimental methodology is also important in order to support replication of results.

Discussion topics when setting up an experimental design

An experimental design or randomized clinical trial requires careful consideration of several factors before actually doing the experiment. An experimental design is the laying out of a detailed experimental plan in advance of doing the experiment. Some of the following topics have already been discussed in the principles of experimental design section:

• How many factors does the design have, and are the levels of these factors fixed or random?
• Are control conditions needed, and what should they be?
• Manipulation checks; did the manipulation really work?
• What are the background variables?
• What is the sample size. How many units must be collected for the experiment to be generalisable and have enough power?
• What is the relevance of interactions between factors?
• What is the influence of delayed effects of substantive factors on outcomes?
• How do response shifts affect self-report measures?
• How feasible is repeated administration of the same measurement instruments to the same units at different occasions, with a post-test and follow-up tests?
• What about using a proxy pretest?
• Are there lurking variables?
• Should the client/patient, researcher or even the analyst of the data be blind to conditions?
• What is the feasibility of subsequent application of different conditions to the same units?
• How many of each control and noise factors should be taken into account?

The independent variable of a study often has many levels or different groups. In a true experiment, researchers can have an experimental group, which is where their intervention testing the hypothesis is implemented, and a control group, which has all the same element as the experimental group, without the interventional element. Thus, when everything else except for one intervention is held constant, researchers can certify with some certainty that this one element is what caused the observed change. In some instances, having a control group is not ethical. This is sometimes solved using two different experimental groups. In some cases, independent variables cannot be manipulated, for example when testing the difference between two groups who have a different disease, or testing the difference between genders (obviously variables that would be hard or unethical to assign participants to). In these cases, a quasi-experimental design may be used.

Causal attributions

In the pure experimental design, the independent (predictor) variable is manipulated by the researcher - that is - every participant of the research is chosen randomly from the population, and each participant chosen is assigned randomly to conditions of the independent variable. Only when this is done is it possible to certify with high probability that the reason for the differences in the outcome variables are caused by the different conditions. Therefore, researchers should choose the experimental design over other design types whenever possible. However, the nature of the independent variable does not always allow for manipulation. In those cases, researchers must be aware of not certifying about causal attribution when their design doesn't allow for it. For example, in observational designs, participants are not assigned randomly to conditions, and so if there are differences found in outcome variables between conditions, it is likely that there is something other than the differences between the conditions that causes the differences in outcomes, that is - a third variable. The same goes for studies with correlational design. (Adér & Mellenbergh, 2008).

Statistical control

It is best that a process be in reasonable statistical control prior to conducting designed experiments. When this is not possible, proper blocking, replication, and randomization allow for the careful conduct of designed experiments. To control for nuisance variables, researchers institute control checks as additional measures. Investigators should ensure that uncontrolled influences (e.g., source credibility perception) do not skew the findings of the study. A manipulation check is one example of a control check. Manipulation checks allow investigators to isolate the chief variables to strengthen support that these variables are operating as planned.

One of the most important requirements of experimental research designs is the necessity of eliminating the effects of spurious, intervening, and antecedent variables. In the most basic model, cause (X) leads to effect (Y). But there could be a third variable (Z) that influences (Y), and X might not be the true cause at all. Z is said to be a spurious variable and must be controlled for. The same is true for intervening variables (a variable in between the supposed cause (X) and the effect (Y)), and anteceding variables (a variable prior to the supposed cause (X) that is the true cause). When a third variable is involved and has not been controlled for, the relation is said to be a zero order relationship. In most practical applications of experimental research designs there are several causes (X1, X2, X3). In most designs, only one of these causes is manipulated at a time.

Experimental designs after Fisher

Some efficient designs for estimating several main effects were found independently and in near succession by Raj Chandra Bose and K. Kishen in 1940 at the Indian Statistical Institute, but remained little known until the Plackett–Burman designs were published in Biometrika in 1946. About the same time, C. R. Rao introduced the concepts of orthogonal arrays as experimental designs. This concept played a central role in the development of Taguchi methods by Genichi Taguchi, which took place during his visit to Indian Statistical Institute in early 1950s. His methods were successfully applied and adopted by Japanese and Indian industries and subsequently were also embraced by US industry albeit with some reservations.

In 1950, Gertrude Mary Cox and William Gemmell Cochran published the book Experimental Designs, which became the major reference work on the design of experiments for statisticians for years afterwards.

Developments of the theory of linear models have encompassed and surpassed the cases that concerned early writers. Today, the theory rests on advanced topics in linear algebra, algebra and combinatorics.

As with other branches of statistics, experimental design is pursued using both frequentist and Bayesian approaches: In evaluating statistical procedures like experimental designs, frequentist statistics studies the sampling distribution while Bayesian statistics updates a probability distribution on the parameter space.

Some important contributors to the field of experimental designs are C. S. Peirce, R. A. Fisher, F. Yates, C. R. Rao, R. C. Bose, J. N. Srivastava, Shrikhande S. S., D. Raghavarao, W. G. Cochran, O. Kempthorne, W. T. Federer, V. V. Fedorov, A. S. Hedayat, J. A. Nelder, R. A. Bailey, J. Kiefer, W. J. Studden, A. Pázman, F. Pukelsheim, D. R. Cox, H. P. Wynn, A. C. Atkinson, G. E. P. Box and G. Taguchi. The textbooks of D. Montgomery, R. Myers, and G. Box/W. Hunter/J.S. Hunter have reached generations of students and practitioners.

Some discussion of experimental design in the context of system identification (model building for static or dynamic models) is given in  and.

Human participant constraints

Laws and ethical considerations preclude some carefully designed experiments with human subjects. Legal constraints are dependent on jurisdiction. Constraints may involve institutional review boards, informed consent and confidentiality affecting both clinical (medical) trials and behavioral and social science experiments. In the field of toxicology, for example, experimentation is performed on laboratory animals with the goal of defining safe exposure limits for humans. Balancing the constraints are views from the medical field. Regarding the randomization of patients, "... if no one knows which therapy is better, there is no ethical imperative to use one therapy or another." (p 380) Regarding experimental design, "...it is clearly not ethical to place subjects at risk to collect data in a poorly designed study when this situation can be easily avoided...". (p 393)