Skip to main content

The Demarcation Problem, Pseudoscience, and Science

  • Author:
  • Updated date:

Mattja is a writer and thinker who likes helping others explore the big questions in life.

This article provides an in-depth discussion about the demarcation problem and what separates science from pseudoscience.

This article provides an in-depth discussion about the demarcation problem and what separates science from pseudoscience.

How Do We Define Science?

Laudan (1983) went so far as to claim that there is no demarcation problem, as in, he believes it is a pseudo-problem to try to determine whether there exists a cleavage between science and non-science, and pseudoscience and science.

This was based on his thinking that the demarcation problem was ill-defined and no coherent demarcation criteria could be provided. He saw any attempts to circumscribe pseudoscience from science always fail.

If astrology can be falsified but so can astronomy, which one is a science? If string theory cannot be falsified and nor can Freud's psychoanalysis, which one is a science? If a psychologist lacks consistent definitions, such as that for "happiness", how can a body of science be built on top of such shaky grounds? If there are no universal, inviolable laws that govern the social sciences, how can these sciences also call themselves "scientific"?

Walsh (2009) examined these questions closely, concluding:

It is important to recognise that Laudan does not claim that science doesn't exist. He agrees that the terms "science" and "non-science" identify a genuine distinction, but he argues that this distinction has no philosophical and epistemological significance [...] Laudan favours the diagnosis [that] there is no solution to the demarcation problem – that's the reason we haven't found one, arguing that the demarcation problem hasn't been solved because it is unsolvable.

Since Laudan called the demarcation [problem] a pseudo-problem, we should direct our efforts to "identify theories that are well-confirmed. We can (and should) evaluate confirmation without considering scientific status" (Walsh, 2009).

Pigliucci (2013) provided a belated response to Laudan. He proposes that we should think of the word science much as we do think of the word game. In the Wittgensteinian sense, a game has no universal definition (Biletzki et al., 2016).

We can think of things that are game-like, are games, or rules of specific groups of games, etc., but generalising to all games encompassing all the nuances of what rules are, what the objectives of games are, and so forth, is impossible.

Precisely as the word science also does not have a generalisable universal definition, even if it seems at first glance that it should, or that we should just trust a lexicographer when he or she tells us what science or a game is.

What we are left with are "family resemblances" of definitions for the word science, rather than any clear-cut definitions for words, which is how Wittgenstein thought about language.

Wittgenstein thought that all of human language was a "language game" and that definitions for words form "family resemblances" with each other rather than there existing any clear-cut definitions for words.

Wittgenstein thought that all of human language was a "language game" and that definitions for words form "family resemblances" with each other rather than there existing any clear-cut definitions for words.

Scientific Laws as a Criterion for Science

In evolutionary biology, there are no laws of evolution, which tell you exactly when a species will speciate, have a mutation become dominant in the population, go extinct, or, on the macro-level, when an entire ecosystem will collapse due to evolutionary pressures, given certain inputs and causal circumstances. Or even what makes a trait evolutionarily advantageous in all instances outside of the fact that it allowed that species to propagate its genes. This is one of the only seemingly inviolable conditions for the evolution of a species.

Survival and passing on genes are the only imperatives in evolution. But what makes something conducive or more evolutionarily adapted will vary infinitely with the complex environment the species is in.

What is the consistent definition of an evolutionary advantage in phenomena such as echolocation for bats, heat-sensitive vision for some snakes, lengthy sleep cycles for sloths, and months of hibernation of certain insects, other than these are conducive to survival and gene propagation? Which is a somewhat tautological argument.

Traits of a species that are selected for by evolutionary pressures were the traits that were required for survival and gene propagation, but we can't say these traits have much else that is necessarily required by evolution beyond that.

What makes one species more adapted than another seems highly random, if you observe the biodiversity of species on earth past and present you see that the variation is mind-boggling. How and why something evolves by natural selection isn't governed, in this sense, by any inviolable laws, only a definite process is occurring where the genes best suited to the surrounding environment and those randomly, naturally, or sexually selected for are passed on to the next generation.

Evolutionary biologists also struggle to define species as there is usually an exception to the rule regarding the taxonomic classification. For example, not all species that cannot reproduce with each other are separate species. Some separate species can create hybrid species that produce fertile offspring (it's likely this occurred with neanderthals and anatomically modern humans), and some plants do not reproduce sexually, but we separate different plant species without using this criterion.

Gene propagation and survival must occur for the evolution of a species to be successful and this may be the closest to a 'law' of Darwinian evolution that there is. However, the same could be argued that the 'law' of 'scientific history' is that time progresses linearly (Berlin, 1960), and humans are causally bound to this law, just like any other law of nature.

Once more, which do we call science: historiography or evolutionary biology? Neither of these notions of scientific law has the same type of mathematical precision and power that other laws such as Newton's laws or Boyle's law or the laws of thermodynamics do or other laws found within chemistry and physics.

Furthermore, the Stanford Encyclopedia of Philosophy's article "Evolution" attempts to give an extensive definition of evolution:

[biological evolution] is change in the properties of groups of organisms over the course of generations…it embraces everything from slight changes in the proportions of different forms of a gene within a population to the alterations that led from the earliest organism to dinosaurs, bees, oaks, and humans.

There is little in such statements which would indicate law-like inviolability. This has been explored by Murray (2001):

Biology has many inductive generalizations (e.g. Bergmann's Rule and 'all cells arise from preexisting cells') but few, if any, recognized universal laws and virtually no deductive theory. Many biologists and philosophers of biology have agreed that predictive theory is inappropriate in biology, which is said to be more complex than physics, and that one can have nonpredictive explanations, such as the neo-Darwinian Theory of Evolution by Natural Selection. Other philosophers dismiss nonpredictive, explanatory theories, including evolutionary 'theory', as metaphysics.

Laws in biological science may include Mendelian inheritance, the Hardy–Weinberg principle, and so on. However, from an article in Scientific American based on the September 23, 1999, lecture that Ernst Mayr, one of the towering figures in the history of evolutionary biology, delivered in Stockholm on receiving the Crafoord Prize from the Royal Swedish Academy of Science:

Laws give way to concepts in Darwinism. In the physical sciences, as a rule, theories are based on laws; for example, the laws of motion led to the theory of gravitation. In evolutionary biology, however, theories are largely based on concepts such as competition, female choice, selection, succession and dominance. These biological concepts, and the theories based on them, cannot be reduced to the laws and theories of the physical sciences [...] Many biologists and philosophers deny the existence of universal laws in biology and suggest that all regularities be stated in probabilistic terms, as nearly all so-called biological laws have exceptions. Philosopher of science Karl Popper’s famous test of falsification therefore cannot be applied in these cases.

It's hard to see that there are any laws of evolution, where mathematical relationships can be formulated and precise calculations and predictions can be made based on input variables and measurement data in an experimental setting. This just cannot happen in evolutionary science, and arguably in biology as a discipline (unless a biologist appeals to underlying biochemical laws for example), even though we can get a probabilistic idea and form hypotheses about what the path of a species will be given certain environmental pressures, we can't produce the type of certainty that is present in physical and chemical laws.

Such an instance is what has occurred in the longest-running evolution experiment, conducted on E. coli to test how this species of bacteria responds and evolves given environmental manipulations in a lab setting. Even knowing the necessary and sufficient conditions, and mathematical formulation of evolution occurring through the Hardy–Weinberg principle, for instance, predicting the future trajectory of the experiment with the highest degrees of probability was not possible.

In fact, researchers were surprised to discover that there does not seem to be a maximum point where a species will stop evolving even when its environment is mostly static. Something only revealed through experiment, and not predicted by the laws that are supposed to govern evolution by natural selection known previously.

Lenski's long-term evolution experiment with E. coli has seen over 50,000 new generations since its inception in 1998.

Mutations in evolutionary history have occurred for a myriad of reasons, and there is usually a species that violates what has been observed in the past regarding what is considered evolutionarily 'advantageous' in one species but not in another. Hence, evolution by natural selection is an explanatory theory seeking to explain why and how life evolved on earth, which has been confirmed by scientists who have tested the claims of Darwinian evolution.

It is a process that takes place where we know very little about predicting precisely how it will play out, even though scientists have closely examined the history of the earth, the fossil record, etc., over billions of years and have an abundance of data regarding the process of evolution of life on earth. Ecosystems and living systems are chaotic in nature and too complex to build models of and accurately predict the future of these systems.

The evolution of human consciousness is an example of the complexity that gave rise to life on earth. Simulating the evolution of human consciousness on a computer, for example, is simply impossible at this point in time and may always be. The evolution of human consciousness did occur, but discerning any scientific laws that underpin it may be in many ways a futile task, save for the chemical and physical laws that biota are causally connected to.

It's not to say that we aren't observing something factual and empirically true about nature and the way it works, it's just our 'laws' and theories regarding evolutionary biology are not suited for predicting the future with high degrees of certainty, which is unlike any other laws in science that have very high levels of predictive power (they are almost certain and absolute and have not been violated after many human experiments to falsify them, but they are also fallible as they can never be absolutely true). Therefore it's best to think of evolutionary theory as a scientific fact rather than a scientific law.

Newton's famous law of gravity, which describes the inverse square relationship between the mass of two objects and the distance between them, determining the magnitude of gravitational force.

Newton's famous law of gravity, which describes the inverse square relationship between the mass of two objects and the distance between them, determining the magnitude of gravitational force.

Certainty in Science

There is, therefore, no way to predict some events studied by what most currently consider sciences (the hard sciences and natural sciences) with high degrees of accuracy, such as how a climate scientist cannot predict the future with much certainty, only giving confidence intervals and probabilities.

And to a further degree, and to serve as the most striking counterexample to certainty in the hard sciences, nor can a physicist tell us when an atom will emit energy due to radioactive decay, or what the position and spin of a particle are at any one time and instant, only the probability of where it will be and what its spin will be, with the more certain one measure is, the less uncertain the other becomes (the Heisenberg uncertainty principle). This is hardly precision of the highest order which is argued for by those who support the point of view that only the hard sciences are real sciences.

Yes, there are dangers in classifying everything as a potential science; however, the requirement that only sciences with immutable laws and almost certain predictive power (or as was argued by Aristotle) universal knowledge and truth gained through inductive reasoning (William, 1922)) used to model physical phenomena, such as Newton's laws, general relativity, chemical reactions, and thermodynamics is too restrictive.

Some areas of study are more scientific than others (Pigliucci, 2013) and within each domain of science there are degrees of usage of scientific methodology; for example, aspects of neuroscience and neurobiology in psychology are more scientific than other aspects of psychology, which include clinical psychology or psychoanalysis.

ESP, Freudianism, parapsychology, flat-earthism, creationism, and intelligent design are barely scientific, with little-to-no empirical and theoretical coherence. String theory, evolutionary psychology, and scientific history have varying levels of theoretical knowledge based on little-to-no experimental confirmation as the experimental methods to empirically test these theories are not known at this time with much confidence if any means to do so exist at all.

The scientific method includes hypothesis testing, statistical methods, experimental evidence, and an incorporation of techniques from other sciences which have a firm footing, this being the "hard sciences". The softer sciences: economics, psychology, anthropology, sociology, etc., gain their scientific credibility from the heavy usage of statistics and empirical testing.

Pigliucci (2013) created a chart to help us think about the different levels of scientific knowledge. Pseudoscience is on the bottom left and the most certain or scientific is the top right.

Pigliucci (2013) created a chart to help us think about the different levels of scientific knowledge. Pseudoscience is on the bottom left and the most certain or scientific is the top right.

Statistics Used as a Means to Make Social Science Scientific

Statistics is an applied science and it is applied mathematics. From the SEP article "Scientific Objectivity":

Moreover, the measurement and quantification of characteristics of scientific interest is only half of the story. We also want to describe relations between the quantities and make inferences using statistical analysis. Statistics thus helps to quantify further aspects of scientific work (Reiss et al., 2017).

The use of statistical techniques such as hypothesis testing, controlling for variables appropriately, and isolating dependent and independent variables is not a trivial task. The achievement of sound statistical studies is based on advanced mathematics and computation, empirical evidence, engineering and scientific techniques.

Claims such as you can make statistics conclude anything (Huff, 1954), is true to a degree. It's true in the sense that poorly designed experiments and statistical studies will necessarily lead to dubious conclusions. However, just because poor statistical studies exist does not mean statistical science and sciences that make heavy use of statistics are invalid. To do so might not matter to many who don't care whether they are called scientists or not.

But to claim the soft sciences and those that employ the heavy use of statistics are not scientific in any way opens the door to those who want to beg the question of how we should instead approach solutions to problems that the soft sciences and sciences that use statistics explore.

As an aside, even the deterministic sciences have chaos built into them and make heavy use of statistics, as I previously mentioned quantum physics does, but others also do, such as statistical mechanics and chaos theory in fluid dynamics (Sommerer et al., 1997). So either we accept that statistics is one of our best tools to help us understand reality through science, or we do not accept the truth, whether it's a high-degree or low-degree of truth, established by theories based on statistical methods.

The Lorenz attractor has deterministic boundary conditions but follows a chaotic and completely random path. This is the nature of chaos theory which is used to model nonlinear systems and phenomena such as fluids, gases, ecosystems, and economies.

The Lorenz attractor has deterministic boundary conditions but follows a chaotic and completely random path. This is the nature of chaos theory which is used to model nonlinear systems and phenomena such as fluids, gases, ecosystems, and economies.

The "Science of Man"

So if the soft sciences aren't really science, then we should not accept that the conclusions they make are representative of reality and instead give more power to philosophers to make purely rationalistic a priori and idealistic explanations of human behaviour.

We could have a cadre of Nietzsche scholars or Hegelian phenomenologists to deconstruct reality for us and do away with scientific truth, especially of the kind purported by social scientists and psychologists. That's not to say Nietzsche or Hegel don't have their value. Just, one who is undertaking a search for truth about reality should not be dismissive of and incredulous about the conclusions that science has revealed to us.

Nietzsche and Hegel are key figures in continental philosophy and postmodern philosophy, and it is no surprise to continental philosophers that this tradition in philosophy takes a mostly anti-science approach to discovering truth.

It's an old dogma that the "science of man" is an endeavour that is disallowed and heretical, with any attempts at creating one being against the sacrosanct purity of God-given nature, or at the very least antagonistic and in conflict with the pursuit of religious worship, claims and behaviour (Shepherd, 1972).

Many who disdain those who make use of science outside of the hard sciences, run the risk of having little understanding of what it is they are criticising, preferring to dismiss anything not falling under the faculty of science proper at a university (famous examples include Richard Feynman), or simply prefer armchair theorising about human nature and how it is idealistic and we couldn't possibly understand it through empirical means. Only pure philosophy and metaphysics of the highest order will save us.

We are, on the contrary, beginning to get an understanding of human nature through the social sciences, and making significant strides to answer seemingly intractable philosophical and scientific questions, such as by using knowledge gained from psychology, neuroscience, neurobiology, and cognitive science (Thagard, 2014), and not so useless are the less experimental sciences (which are becoming less-so with time such as economics (Rosenzweig et al., 2000), sociology, and political science.

Of course, these disciplines are not without their limitations, and, for instance, we are beginning to better understand, through cognitive science, philosophical notions such as innateness, meaning, folk psychology, mental states, moral psychology, free will, emotions, mental illness, and even the meaning of life.

Cognitive science may not effectively address or cannot address questions about human nature, such as whether human thought is more computational or dynamical, whether consciousness can be understood through a scientific lens and the vast complexities of human social interaction. And other areas of science may possibly help philosophers in those areas, for example, by using knowledge of physics, political science, economics, and sociology, or, perhaps, these are problems which can never be dissolved using any scientific means.

Scientific Theories of Human Nature, the Fallibility of Scientific Knowledge, and Postmodern and Neopragmatist Responses to Scientific Knowledge

Theories about nature and human nature are bound to be wrong. Just as in the past when Galileo challenged the Catholic church's views of a geocentric universe that pulled all matter towards the centre of the earth, Einstein challenged Newton, Darwin challenged the science of the day, and how string theorists now challenge the limits of the standard model in physics, we have often been wrong and will continue to be wrong about our notions of reality when new scientific evidence is revealed to us. What's most important, however, is how scientific our search for knowledge is.

Laudan was right that there may not be a universal definition of science or pseudoscience; however, this is not necessary for doing science. There are degrees of scientific knowledge, just as there are degrees of meaning for different definitions of the word game. We know the word science when we hear it or read it, and we recognise it like when we recognise the similar physical characteristics of related family members. We can see the similarity between cousins or brothers, but we, on the other hand, do not see the same similarities between complete strangers. This is analogous to the contrast between pseudoscience and science, where pseudoscience is a complete stranger to science.

But to say the word science or the demarcation between science and pseudoscience is utterly meaningless, as Laudan may have gone so far to claim, or at least be interpreted as claiming, opens the door to many undesirable epistemic vexations. Laudan's arguments are relevant to discussions of creationists who have tried to justify teaching 'creation science' in high schools in US courts, such as the McLean v. Arkansas case, in 1981, where the court determined creationism to be a pseudoscience and not to be taught in public schools (Ruse, 1982). Although not a creationist himself, and a supporter of establishing evolutionary theory as scientific, according to Ruse (2018),

Laudan and company inspired the Creationists to new efforts, and since the Arkansas court case, the philosophical dimension to the evolution/Creationism controversy has been much increased. In particular, philosophical arguments are central to the thinking of the leader of today's creationists, Berkeley law professor, Phillip Johnson, whose reputation was made with the anti-evolutionary tract Darwin on Trial.

Those who argue that since we cannot unequivocally and universally state what pseudoscience means, therefore, distinguishing science from non-science or pseudoscience is an impossible task, appear to be using a postmodern sleight-of-hand, and game, with words that take the philosopher Wittgenstein in a direction he may not have been happy to accept: a world devoid completely of meaning. If science is our most important tool for establishing approximate truth about the world, and we can't agree upon what is science and is not science due to semantic quibbles, what hope do we have for knowing much of anything about reality through science outside of only the hard sciences?

The later Wittgenstein was radically different from the earlier, but one who is familiar with his later work, and has studied it closely, shouldn't get the impression that Wittgenstein thought intersubjective meanings were impossible. Perhaps some, chiefly postmodernists, would interpret him that way. Using Wittgenstein as ammunition to discredit even all of science, where truth is only the truth when collectively we construct it to be so. Postmodern social constructivists hold this position about science, as pointed out by Goldman et al. (2016):

some sociological approaches to science claim to show that scientific “facts” are not “out-there” entities, but are mere “fabrications” resulting from social interactions. This metaphysical thesis is a form of social constructivism.

Even neopragmatists such as Rorty, have been accused of this type of radical relativism.

Rorty wrote in Objectivity, Relativism, and Truth: Philosophical Papers,

there is nothing to be said about either truth or rationality apart from descriptions of the familiar procedures of justification which a given society—ours—uses in one or another area of inquiry. (Rorty 1991: 23)

Therefore, you can choose the postmodernist camp or radical relativist camp that some neopragmatists appear to endorse, but you must then accept that coherent meaning is impossible between individuals, even if you have agreed upon definitions, the truth will only be dependent on consensus, it's not "out there" it's not mind-independent, it depends on our constructions of it.

The philosophy of language is central to helping define what science and non-science and pseudoscience are. For in-depth, academic, and professional studies of nature, the word science is clearly enough defined for pragmatic purposes, to achieve what scientists and philosophers of science set out to do. That being, to clarify what we mean when we talk about nature, what it consists of, and how it works, based on painstaking evidence collection, experiment, and research, employing the best tools: mathematical, scientific, or otherwise to understand what nature is like.

Geocentrism was a dogma of Galileo's time, which he challenged and was forced to recant his views later under the orders of the Catholic church.

Geocentrism was a dogma of Galileo's time, which he challenged and was forced to recant his views later under the orders of the Catholic church.

What Science Ought to Be About

The scientific enterprise is about explaining how nature works using our best methods. Science is not reporting about events, creating beauty, used for entertaining idle minds, or used by those who can speak science-lingo to confuse, confound, and bamboozle those who aren't well-versed in science-speak. Those things may be elements and consequences of the practice of science to some but not the primary concern of a scientist, at all, in their domain of expertise.

An approximation of the true nature of reality is what a scientist ought to study. This approximation must be based on reality and it cannot be purely based on theory without any anchorage to empirical confirmations or well-grounded empirical and scientific knowledge, and it cannot be grounded in fantasy and wishful thinking.

One who has a poor understanding of science and logic, and falls victim to multifarious human biases is a cancer that infects and causes poor reasoning, misinformation, misunderstanding, and pseudoscience. There's no better word for human enquiries such as astrology, creationism, and alchemy than pseudoscience, now that we know better as a species.

This distinction between science and pseudoscience does differ from non-science to science. Non-science is when science is done, but it's wrong, empirically flawed rather than theoretically or experimentally dubious, etc., such as when data is incorrectly tabulated, measurements aren't correctly gathered, and human error causes other mistakes in applying scientific methodology, and rather than when scientific methodology is flawed, disproven, and defunct to begin with (which is pseudoscience).

I, therefore, argue strongly for the continuation, rather than sanitisation, of the usage of the word pseudoscience; otherwise, we will have no power over our language and whatever truth we want to be will be, and the goal of objectivity will become nothing more than a hindrance, setting the clock of history turning in reverse, towards the dark ages.


Berlin, Isaiah (1960). History and Theory: The Concept of Scientific History. _History and Theory_ 1 (1):1.

Biletzki, Anat and Matar, Anat (2016). "Ludwig Wittgenstein", The Stanford Encyclopedia of Philosophy (Fall 2016 Edition), Edward N. Zalta (ed.), URL = <>.

Goldman, Alvin (2016) and Blanchard, Thomas. "Social Epistemology", The Stanford Encyclopedia of Philosophy (Winter 2016 Edition), Edward N. Zalta (ed.), URL = <>.

Hansson, Sven Ove (2017). "Science and Pseudo-Science", The Stanford Encyclopedia of Philosophy (Summer 2017 Edition), Edward N. Zalta (ed.), URL = <>.

Huff, Darrell (1954). How to Lie with Statistics (illust. I. Geis), Norton, New York,

Laudan L. (1983). The Demise of the Demarcation Problem. In: Cohen R.S., Laudan L. (eds) Physics, Philosophy and Psychoanalysis. Boston Studies in the Philosophy of Science, vol 76. Springer, Dordrecht

Millstein, Roberta L. (2017). "Evolution", The Stanford Encyclopedia of Philosophy (Fall 2017 Edition), Edward N. Zalta (ed.), URL = <>.

Pigliucci, Massimo (2013). The demarcation problem: a (belated) response to Laudan. In Massimo Pigliucci & Maarten Boudry (eds.), _Philosophy of Pseudoscience: Reconsidering the Demarcation Problem_. University of Chicago Press. pp. 9.

Reiss, Julian and Sprenger (2017). "Scientific Objectivity", The Stanford Encyclopedia of Philosophy (Winter 2017 Edition), Edward N. Zalta (ed.), URL = <>.

Rosenzweig, Mark R. and Wolpin, Kenneth I. (2000). "Natural 'Natural Experiments' in Economics", Journal of Economic Literature, Vol. 38, No. 4 (Dec., 2000), pp. 827-874

Rorty, Richard (1991). Objectivity, Relativism, and Truth: Philosophical Papers, Vol. 1, Cambridge: Cambridge University Press.

Ruse, Michael (1982). "Creation science is not science", Science, Technology, and Human Values 7, no. 40 pp: 72-78

Ruse, Michael (2018). "Creationism", The Stanford Encyclopedia of Philosophy (Spring 2018 Edition), Edward N. Zalta (ed.), forthcoming URL = <>.

Shepherd, W. (1972). Religion and the Social Sciences: Conflict or Reconciliation? Journal for the Scientific Study of Religion,11(3), 230-239. doi:10.2307/1384547

Sommerer, John C., Edward Ott, and Tamás Tél (1997). "Modeling Two-Dimensional Fluid Flows with Chaos Theory", JOHNS HOPKINS APL TECHNICAL DIGEST, VOLUME 18, NUMBER 2 (1997) 193

Thagard, Paul (2014). "Cognitive Science", The Stanford Encyclopedia of Philosophy (Fall 2014 Edition), Edward N. Zalta (ed.), URL = <>.

Walsh, K. (2009). Has Laudan killed the demarcation problem? Masters Research thesis, Arts - School of Philosophy, Anthropology and Social Inquiry, The University of Melbourne.

William M. Dickie (1922). A Comparison of the Scientific Method and Achievement of Aristotle, The Philosophical Review, Vol. 31, No. 5 (Sep., 1922), pp. 471-494 Published by: Duke University Press on behalf of Philosophical Review Stable URL: Accessed: 10-03-2018 21:52 UTC

© 2018 Mattja