Philosophy of Science (Course Notes and exam prep.)

[I took a course in the philosophy of science this year and although the lecturers and professor are nice people – I did not like overall experience. I felt there was very little philosophy being done, more just a critical review of science and the information that it has generated. I am surprised that I passed this exam because I did not feel good writing my essay’s in those two hours… Yet, like I said the Lecturers and professors are good but the information contained in the curriculum needs a lot of work if it is to become a more enjoyable learning experience. Anyway, I wish the professors and philosophers of Science in Leuven the best of luck in this regard. I have collected the contents of my preparation for the exam so I do not loose the notes and can review at a later date. Perhaps they may be of interest to some…]


Construct an argument that either defends or critiques a statement below.

896944195679_Digimedia_EinsteinBirthdayProject_DELIVERY_TXTD.00_00_53_12.Still001 (1)

  1. Philosophy of science is useless for the practice of science.

The science we know today would not have come into existence if there had not been what was widely called ‘natural philosophy’. A way of scientific inquiry which contained a process of questioning the world in its material nature. This has a very long history indeed the western centric textbook answer has it that this strand of philosophy that was ‘natural philosophy’ began with Aristotle (The Reader, as Plato called him) who was one of the first to systematically record his scientific explorations of the biological world. Then, we have been told to believe that the transformation happened under the guidance of Rene Descartes the father of Cartesianism. A dualistic belief that the body and mind are separate substances one is extended the other thinking. But, at this point in history science remained in the nurturing womb of philosophy and had yet to branch off into its current form.

This happened when towering figures such as Copernicus, Galileo, and Isac Newton showed how by using maths a human might offer an explanation for the way the world works to such accuracy that knowledge became wedded to science. Of course, this had an effect on philosophy no longer the sole proprietor of Truth. Being forced to observe this baby called science philosophy held it as an object of its reflections. One such example being the creator of Positivism the thinker Auguste Comte who wrote about how science was an inevitability in his somewhat Hegel inspired tripartite movement from metaphysics and theology to positivism construed as a study of the relations between a growing body of natural knowledge; such a system is pro-science, ‘When these different operations are sufficiently advanced to have assumed an irrevocable character, we shall see social education itself fall for ever into the hands of the scientists.’ (Comte, …) Words like this contain what many would like to believe, that science is a social education, progressing by way of shared research. There is no longer the need for individual greatness or of the speculative prolonged reflections philosophy offers – science now has a track record of knowledge production.

Such a situation where science has buried its parents might sometimes feel like it is the case or even desirable: in 2010 Stephen Hawking went on record as saying he believes philosophy to be dead. So, this leaves us at a contested position either philosophy still has something to add to the sciences or it is indeed useless? However, this misses an important part of the question: ‘useless for the practice of science’, so it is of course essential to leave behind the debate on the qualities of the two separate disciplines and focus on what they each have to say about one another’s practices. Philosopher of Science Samir Okasha explains the practice of science to contain two distinct features the experiment and the theory. Okasha hones in on the valuable role he thinks philosophy of science performs; it continues questioning when the scientist equipped with belief in the reproducibility of his experiment stops questioning. He mentions this in line with a problem science and philosophy both share how to differentiate between that which is pseudo- or just operating under the name of the practice.

I would argue that considering all of what one has just mentioned philosophy of science remains very useful for the practices of science. For the following reasons should suffice to support this: 1) Science often needs more ethical considerations – just think of the Manhatten project, nuclear power, and the future of DNA editing, 2) Occasionally science has discovered something remarkable but it may not know how to communicate precisely what the discovery is telling them – a perfect example would be quantum physics and the human mind. 3) Philosophy exerts parental rights over science as its history shows it was rationalist philosophers such as Leibniz and Descartes who inspired Newton’s breakthroughs in calculus and physics. A potential problem for this conclusion is that the practices of some philosophers who may habitually over question, or harbour questions that are unfairly weighted against science containing a prejudice will inevitably be disruptive. Yet, I would suggest that such a philosopher has not spent enough time reflecting and therefore in the rare event that they come into contact with the practice of science one does not imagine that the scientist would put up with such distractions for long. The probability of this is very low and controlled and well thought out criticism of both practices is obviously encouraged.

Finally, one last argument in favour of the usefulness of the philosophy of science:

  1. Let us say that a progressive scientific research programme is near a new discovery in Dark matter and energy.
  2. The scientific experiments provide certifiable and undeniable evidence for the But, the scientific community does not know what their results, the discovery actually means outside the context of the experiment. It refutes earlier theories but leaves questions unanswered.
  3. Therefore, science will greatly benefit from having a philosopher who used to questioning questions will bring new abstract interpretations to the table, and thus greatly improve the probability of agreeing upon what the discovery means and how it should be communicated.

  1. A better interpretation of probability could solve the problem of induction.

Let us begin with a clear definition of what induction is and why it has for so long been considered problematic. Induction is a form of argument that stands in contrast to deduction. Deduction is an argument that contains a conclusion deduced from its premises this makes it necessarily true because it follows that if the premises of the argument are all true then the conclusion must be true. Abduction, differs from this because the argument contains premises/inferences that do not necessarily support the conclusion an example of such abduction and its reasoning is below:

A large quantity of reports contain errors of calculation

All the reports were written by the same person.

Therefore, all the reports will have to be re-written.

Here it is evident that if we compare the first premise with the conclusion it follows that it is only contingently true that all the reports will have to be re-written because a large quantity is not ‘all reports’. Following Okasha we can understand why this way of reasoning is an issue for philosophy in that inductive reasoning is often found in everyday use. Okasha’s examples include the following, ‘when you turn the steering wheel of your car you assume that it will turn the way it turns because of past events’(Okasha…). Continuing thinking through inductive reasoning Okasha cites the Scottish Enlightenment thinker David Hume  who was the first to offer up an explanation for the dominance of inductive reasoning in our everyday experiences. Hume claimed that it was due to pure animal habit that we reason in such way and so when we induce that the sun will rise in the morning tomorrow we do so because of the Uniformity of Nature (U.N)(Hume, ). In short this is an assumption relative to objects we have or have not observed. Hume continued when considering if the U.N could be proven and he denied that it could stating that there could exist a universe where nature was not uniform and existed in a state of constant flux. Essentially, Okasha helps distill Hume’s point that: there is no way of empirically proving the uniformity of nature without trying to persuade someone who does not trust in induction is a process of induction thus committing the formal fallacy of begging the question.

You may say that it seems like one of those stereotypical problems that philosophers bicker over and you would be correct. But, many philosophers argue that induction is so essential to how we think that it is not something provable. Although there are the following responses to problem of induction:

  1. Peter Strawson’s analogy: ‘If someone worried about whether a particular action was legal, they could consult the law-books and compare the action with what the law-books say. But suppose someone worried about whether the law itself was legal. This is an odd worry indeed. For the law is the standard against which the legality of other things is judged, and it makes little sense to enquire whether the standard itself is legal.

Induction is a standard to which we decide whether or not our claims are justified.

  1. Inference to the best explanation (I.B.E)


Basic everyday induction takes the form of:  ‘all x’s examined so far have been y’,

and the conclusion has had the form ‘the next x to be examined will be y’, or sometimes, ‘all x’s are y’. In other words, these inferences take us from examined to unexamined instances of a given kind.

In I.B.E there can not be two events that infer the conclusion so we have a most probable one take the example argument below:

The left over curry in the fridge has been eaten.

The husband/wife arrived home from work late.

The Husband/Wife ate the left over curry.

Charles Darwin used Inductive reasoning in his theory of evolution saying that evolution or the development of species only makes sense if there is an observable relation a common ancestor (horses and zebras for example).

Okasha, explores a potential disagreement with (I.B.E) that it remains uncertain as to how to distinguish between possible explanations and the data present in the argument. The solution is that the explanation that is the best is also the most simple one. Yet, using simplicity and parsimony (…?) as solutions still does not resolve the problem because it does not say anything about the main issue that ‘the uniformity of nature’ makes problematic that the universe may be either simple or complex.

Part of the confusion surrounding how to resolve this is the problem of interpreting the word ‘probability’ some say that when we state the probability of something happening let us say the chances of me cooking a vegetarian dish tonight are 1/10 rather than an exact percentage or prediction it communicates a subjective interpretation.

[contrasting with the usual frequency interpretation of probability: If you read that the probability of an Englishwoman living to 100 years of age is 1 in 10, you would understand this as saying that one-tenth of all Englishwomen live to the age of 100. / But what if you read that the probability of finding life on Mars is 1 in 1,000? Does this mean that one out of every thousand planets in our solar system contains life? Clearly it does not. For one thing, there are only nine planets in our solar system.]

[The logical interpretation of probability rejects the idea that there are no objective facts about probability (subjective interpretation) by saying that there is true and false positions regarding events. Evidence for this Advocates of the logical interpretation think that for any two statements in our language, we can in principle discover the probability of one, given the other as evidence. For example, we might want to discover the probability that there will be an ice age within 10,000 years, given the current rate of global warming.]

{0.9 < maximum is one 1/10,1000 I.e once every ten thousand years}

  • Mendelian genetics, which deals with the transmission of genes from one generation to another in sexually reproducing populations. One of the most important principles of Mendelian genetics is that every gene in an organism has a 50% chance of making it into any one of the organism’s gametes (sperm or egg cells). Hence there is a 50% chance that any gene found in your mother will also be in you, and likewise for the genes in your father. Using this principle and others, geneticists can provide detailed explanations for why particular characteristics (e.g. eye colour) are distributed across the generations of a family in the way that they are. Now ‘chance’ is just another word for probability, so it is obvious that our Mendelian principle makes essential use of the concept of probability.
  1. (Okasha) “For John and Jack both accept the evidence that the sun has risen every day in the past, but Jack fails to realize that this evidence makes it highly probable that the sun will rise tomorrow, while John does realize this. Regarding a statement’s probability as a measure of the evidence in its favour, as the logical interpretation recommends, tallies neatly with our intuitive feeling that the premisses of an inductive inference can make the conclusion highly probable, even if they cannot guarantee its truth.”

As a statement in and by itself yes it is the case that a “better interpretation” may one day solve the problem of uncertainty surrounding inductive reasoning and arguments. However such an interpretation would seem to need to be inhumanely accurate to factor in the relation between chance and uncertainty. Maybe a future quantum computer may make advances in probability that will enable us to resolve this issue, but one remains highly sceptical of such a solution because it would imply the possibility of a world without uncertainty and irrationality and this one believes to be unattainable and undesirable for a science.

  1. Falsificationism rejects confirmation and verification, and thus can resolve the problem of induction.

Karl Popper Science: Conjectures and Refutations

  1. “Mr Turnbull had predicted evil consequences …, and now was doing the best in his power to bring the about the verification of his prophecies.”

-Anthony Trollope

  1. “The problem that troubled me at the time was neither, ‘when is a theory true?’, or ‘ when was a theory acceptable?’. My problem was how to distinguish between pseudo-science and science?”
  • Popper was thrilled with the affirmative experiment and confirmation of Einstein’s calculations for gravity by Eddington’s Eclipse Observations in 1919
  1. Popper had a problem with three theories: Marx’s ‘theory of history’ (Historical Materialism), Freud’s (psychoanalysis – unconscious), and individual psychology. He held a problem with their claims to science because of their lack of certainty or success when measured to the objective predictions they made… compared to the certainty of Einsteinian physics… For Popper these three theories resembled myths rather than science and astrology rather than astronomy.

Popper’s peers are impressed by these theory’s explanatory power, and that they seem to have evidence for their validity everywhere in the world. But this is not the case for Popper.

  1. Against Freud and Adler, Popper used this analogy, ‘Using two choices one human being is confronted with when a man pushes a child into the water with the intention of drowning it, and another that sacrifices his own life to save the youth. Each of the cases can be easily explained via way of Freud and Adler’s theories. In the first instant ( The man suffered from repression/while the second managed sublimation). Secondly, in Adler’s language the first man suffers from inferiority produced due to a need to prove something, the second man is the same he needs to prove he can save the child, The theory’s always seemed to have a conclusive answer and conclusion regardless of the scenario.
  2. He says that Einsteins confirmation of the light of a star during an eclipse bends making it appear further away from the sun. Popper mentions the aspect of risk in these scientific predictions. If they do not match the reality of the world exactly then they are refuted.
  • A) For popper you should not be chasing after confirmations but good scientific theories are a prohibition they forbid certain things to happen. B) Irrefutability is not a good aspect of a theory it is a vice. C) Testability is a way of falsifiability (important). D) Occasionally a scientific theory is saved from refutation because there is an ad hoc auxiliary assumption or hypothesis. But, this destroys the viability and validity of the theory, lowering its status.
  • The Criterion of a scientific theory is its falsifiability, testability, and refutability.
  1. The Criterion of falsifiability is about drawing a line between theories with empirical evidence (science) and those without.
  2. The above is an answer to theproblem of demarcation… “because it says that systems of statements in order to be qualified as scientific… must be capable of conflicting with possible or conceivable observations.

… Imre Lakatos Science and Pseudo-science…


  1. “Blind commitment to a theory is not an intellectual virtue, it is a crime” Scepticism towards one’s own theories is essentially scientific… Belief’s role in formatting knowledge is suspended …
  • Objectivity is essential for science:

‘If we take in our hand any volume; of divinity, or school metaphysics, for

instance; let us ask, does it contain any abstract reasoning concernig quantity

or number? No. Does it contain any experimental reasoning concerning matter

of fact and existence? No. Commit it then to the flames. For it call contain

nothing but sophistry and illusion.’   

  • David Hume, An Inquiry Concerning Human Understanding (1748)
  • Newton once confidently claimed he only produces proposals based upon facts, and especially Kepler’s facts about the movement of the objects in outer space. This was incorrect because Kepler’s facts stated that planets moved in ellipses. Newton claimed that planets would move in ellipses if they did not disturb each other in their motion. However they did so Newton was forced to develop a pertubation theory that states that no planets move in ellipses.
  • ‘inductive logicians’. Inductive logic set out to define the probabilities of different theories according to the available total evidence. If the mathematical probability of a theory is high, it qualifies as scientific; if it is lowor even zero, it is not scientific. Thus the hallmark of scientific honest) would be never to say anything that is not at least highly probab Probabilism has an attractive feature: instead of simply providing a black-and-white distinction between science and pseudoscience, it provides a continuous scale from poor theories with low probability to good theories with high probability. But, in 1934, Karl Popper, one of the most influential of our time, argued that the mathematical probability. of all theories, scientific or pseudoscientific, given any amount of evidence is zero.” If Popper is right, scientific theories are not only equally Inprovable but also equally improbable.
  1. Tom Kuhn, adistinguished American philosopher of science, arrived at this conclusion after discovering the naivety of Popper’s falsificationism. But if Kuhn is right, then there is no explicit demarcation between science and pseudoscience, no distinction between scientific progress and intellectual decay, there is no objective standard of honesty. But what criteria can he then offer to demarcate scientific progress from intellectual degeneration?
  • Now, how do scientific revolutions come about? If we have two rival research programmes, and one is progressing while the other is degenerating, scientists tend to join the progressive programme.

  1. Creationism is a science. (And what is the implication for whether it is rational to believe in creationism?)

Elliot Sober, ‘Creationism’ in Philosophy of Biology, (2000)

  1. Begins by discussing phrenology (measuring the human skull to distinguish behaviours of the mind… as a serious research programme in the past now regressive.
  2. WE must distinguish the people from the propositions they maintain.
  • The earth is flat but this does not stop there being ‘flat-earthers’
  1. Scientific (added to creationism to imply that it appeals to evidance for the existence of god. Creationism vs Evolution / A intelligent being a designer vs natural selection
  2. He assesses the logic both positions defend…
  3. He suggests that creationism has not developed a scientific research programme and still only makes one claim an appeal to God. Evolution on the other hand has grown with many hypothesis tested and grown into a progressive research programme.
  • Sober comments on the authentic intellectual background of the ‘design argument’ explaining that rational theology was a tradition that contained a lot of what was best of western philosophy due to its grounding in reason and rationality.
  • Summa Theologiae, Thomas Aquinas wrote five reasons for the existence of god… the fifth of these arguments is ‘intelligent / argument from design; (1224-1274) building upon ideas developed by Plato and Aristotle.
  1. This argument from design met its heyday with Hume’s Scepticism Dialogues Concerning Natural Religion (1779) and was never believed in in the same way again. (compared to the Bridgewater Treatise)
  2. Abduction Logic of design argument laid out by William Paley, Natural Theology (1805) – An inference to the best explanation containing two possibilities: 1) God is an intelligent designer an engineer so he built organisms that would be well suited to their habitat. 2) Random lumps of matter where transformed by random physical forces into living things. Paley wants to show the former as being more probable.
  3. He uses an analogy of a complex object, a watch with moving parts that functions as a whole. Its success as an object is because it had an intelligent designer.
  • The Likelyhood Principle Edwards 1972.

Consider a statement we know to be true O. Then consider two explanations  (H1…H2...) for why O is true. The likelihood principle reads as follows: O strongly favours H1  over H2 if and only if H1 assigns a higher probability to O than H2 does…

In the notation of probability theory this says:

strongly favors H1 over H2  if and only if P(O/H1) >> P(O/H2).

Expressing the likelihood that hypothesis 1 has in light of observation but don’t confuse

‘It is likely’ and ‘it is probable’ P(O/H1) – P(H1/O) How are they different? Consider the following:

You are sitting in a cabin one night and you hear rumbling in the attic. We wonder what could have produced the noise. I suggest that the explanation is that there are Gremlins in the attic and they are bowling. You dismiss this as implausible. Observation over hypothesis is probable … Hypothesis then observation has a likelyhood but low probability.

Applied to Paleys argument:

A: The watch is intricate and well suited to the task of time keeping.
W1: The watch is the product of intelligent design.
W2: The watch is the process of random physical processes.

Paley claims that P(A/W1) >> P(A/W2) . He then says the same pattern of analysis applies to the following triplets of statements.

B : Living things are intricate and well suited to the task of surviving and reproducing.
L1: Living things are the product of intelligent design
L2:  Living things are the product of random physical processes.

Paley argues that if you agree with him about the watch then you should agree that

P(B/L1) >> P(B/L2) .   

Hume (analogy arguments) stronger or weaker according to how similar the two objects are:  Blood circulates in humans / humans are similar to dogs/plants / dogs plants blood circulates.

Object A has property P

Object A and property T are similar to degree

N _________________________________________

Object T has property P.


N measures the degree of probability two objects are alike n = o / n = 1

For Hume this shows that even as an argument of analogy the degrees of similarity between a living organism and a watch are not enough to make the argument feasible but Paley’s argument may stand alone.

Third use of the likelyhood principle consider we toss a coin a thousand times and note on each toss whether the coin lands heads or tails. We record the observational results in statement O below and wish to use O to discriminate between two hypothesis.

O:  The coin landed heads on 803 tosses and tails on 197
H1:  The coin is biased towards heads – its probability of landing heads when tossed is 0.8
H2:  The coin is fair – its probability of landing heads when tossed is 0.5

  • Inference in induction cause and effect based upon prior knowledge of the probable cause. – Hume

  • Problem with the Design argument and induction: Sample size “Suppose we have good reason to believe that the organisms in our world are the product of intelligent design, then we must have looked at lots of other worlds and observed intelligent designers creating organisms there. We have observed no such worlds so our sample size for postulating the design argument is 0

Darwin – Natural selection and the survival of the fittest… if it involves an element of chance as to the evolutionary selection of species and the survival of hereditary beneficial genetic traits. Then this makes it a random process NO

Natural selection includes unequal probability and for this reason it is not a random process. 1) variation must arise within the population, 2) natural selection goes to work selecting modifying the frequencies of the variants present.

  • Richard Dawkins The Blind Watchmaker (1986)

Imagine a mechanical device that like a combination lock composed of series of disks side by side. On each side the 26 letters of the alphabet are placed. There are 26 possibilities on each disk and 19 discs giving  different possible sequences.

-one of these is: METHINKSITISAWEASEL… the probability of this being spun is 1/  a very small one … but the analogy applies to evolution because the device can be adapted so that when one of the target letters is viewed the device freezes it greatly increasing the speed at which the ordered whole can be attained. Natural selection works in a similar way.

Variation (not about usefulness – more about from the variants what can be retained)   and Retention.


Voltaire, satarized Leibniz’s God created the most perfect world… with DR. Pangloss in Candide


Jacob, natural selection is a tinkerer — argument via similarity / Vestigial Traits (

  • Panda’s boney thumb related to bears…

  • tree of life … common ancestor — all animals share a genealogical history…

DNA( amino acids )  RNA(messenger)

“Coding is arbitrary then it affects the likelihood argument… if the genetic code was the most functionary then we would expect all terrestrial life to use it regardless of origin.”

The problem of productive equivalence …



         O : Organisms are imperfectly adapted to their environment
        Dp: Species were separately created by a super-intelligent and omnipotent god
Who wanted to make organisms/ perfectly suited/adapted to their environment.
 Ev :  Species evolved from common ancestors from natural selection.

The observations are said to favour a hypothesis of evolution  over the perfectionist

Design hypothesis Dp: P(O/Ev) >> P(O/Dp).  But now consider a Trickster (Descartes god as perfect being / trickster)

hypothesis: D1 : Species were created separately by a god who made them look just the way they wood if they had evolved from natural selection.

Ev and DT are predictively equivalent… The Likelihood Principle is baced upon a comparison of competing hypotheses.

Creationism is not a scientific argument because it is un-testable … influenced by Karl Popper (Falsifiability is the hallmark of scientific questioning)


Popper used to believe that evolution was a metaphysical research programme… but changed his mind.

  1. Observation sentences (popper)

Poroposition P is falsifiable if and only if P deductively implies at least one observation sentence O.


Problem: Observation is often theory laden… our perception is not separable from theory.

Poppers Falsifiability Criterion has deeper problems:

1)Tacking Problem

Suppose that some proposition S is falsifiable then it immediately follows that S in a conjunction with another proposition N is also falsifiable. That is if S makes predications that can be checked observationally, then so does S&N. This is an embarrassment for Poppers theory because he wanted to distinguish between the scientific and the non scientific.

  • Strange relationship of a proposition to its negation.

Consider the statement of the form “ All As are B”. Popper judges this as falsifiable because you can observe a single A that is not B. Now consider the negation of the generalization, “There exists an Object that is both A and not – B” This statement is not falsifiable. No single observed object or finite collection of them can falsify the existent claim. Generalization is falsifiable, and the negation is not. Surely is a statement is scientific its negation is falsifiable suggesting that falsifiability is not a good criterion for being scientific.

  • Theories make testable hypothesis when they are conjoined with auxiliary assumptions T by itself does not deductively imply O, but rather T&A.

Peirre Duhems  thesis theory and auxillary hypothesis… Dinosaur and meteor … the theory said nothing about metal iridium being located in certain places so theory needed auxillary hypothesis … this metal has higher concentrations in meteors than found on earth.

  • Probability statements in science are unfalsifiable “ A coin toss is fair because of 0.5 probability” well what if you toss it five times?

[Evolution against Creationism… unscientific main arguments can not be tested

Creationism against Evolution … Scientific theories are often incomplete or are refuted.]

Examples of Poppers problems:


Faslification Verification

If T then O                                                                                 If T, then O

Not –O                                                                                          O

________                                                                                       __________

Then not –T                                                                                 T

(Deductively Valid)                                                                (Invalid)


If T&A, Then O                                                                          If T&A, Then O

Not-O                                                                                           O

_____________                                                                                 ______________

Not – T                                                                                            T

(Deductively Valid)                                                                (Invalid)


A vestige of Poppers asymmetry can be restored if we include the premiss that

The auxiliary assumptions (A) are true…

Falsification                                                                            Verification

If T&A, then O                                                                          If T&A, then O

A                                                                                                  A

. O                                                                                                O

_____________                                                                              _____________

Not- T                                                                                          T

(deductively Valid)                                                                   (Deductively Invalid)

To falsify we have to assume that A is true

Left argument asserts that if we cannot verify theoretical statements, Then we can not falsify them either!!




The Virtue of Vulnerability

Vulnerability appears to be a defect and not a virtue … of science. Why is important that our hypothesis be refutable and vulnerable?

The Liklyhood Principle helps answer these questions. A consequence of this principle is that If  O  favors H1  over H2 , then not-O would favor H2 over H1  .

Because P(O/H1) > P(O/H2) , then P(not – O/H1) < P(not -O/H2) For our beliefs to be supported by observational evidence. For, this to be possible there must be possible observations against them.

“Duheim’s thesis say the hypothesis in science makes testable predictions only when they are conjoined with auxiliary premises / assumptions. Creationists claims that organisms are the result of an intelligent designer is no different. The only distinguishing factor is that creationist auxiliary assumptions are not independently supported. If we can not choose test between auxiliary assumptions then the design hypothesis is not validated.

Sometimes creationists criticize evolutionary biology and philosophy as too naturalistic    

But science is commited to a methodology and not a substantive claim about the way the world should be…

Difference in arguments makes creationism un- falsifiable.



  1. Explanations should be arguments.


In this essay one will provide an argument in support of the statement, ‘explanations should be arguments’. One will do this by citing sources within the practices of the philosophy of science. An area of philosophy which does not seek to think like a scientist even though this often is the case, yet the philosopher who has science as the subject of their thought is faced with a maze of initial questions: the simplest would be what exactly is science? How are we to understand its qualities such as power (political/cultural), importance, and accuracy? Amongst these considerations there are the questions that could be asked surrounding the difficulty of placing or situating explanations and arguments. Both, are essential to science but considering them philosophically the two do not appear to be as clear and distinct as one might initially assume. To overcome the assumption that one understands these two component parts of science I will maintain a simple line of reasoning. Starting with the presumption that if arguments were not explanations than this would make the whole praxis of science a sad unsocial enterprise without its current relevancy.

Such a reality is not true and this is because of the explanatory power of arguments and vice versa the argumentative force of explanations. This is observable in texts by Samir Okasha in his introduction to this strand of philosophy (2016), and David Lewis discussing ‘causal explanation’ (1986). The later text begins with a consideration of an explanandum event, and the causal chain leading up to it ad infinitum. Implying that in the event of describing a phenomenon many causes may be found together or even as part of the explanadum. Lewis articulates in the reductive spirit of science the importance of information in explaining and how this is dependent on a causal history.

‘The why-question concerning a particular event is a request for explanatory

information and hence a request that an act of explaining be performed.

(Lewis, 1986. 218)’

What is forthrightly expressed here is the structural relation between information and an act of explaining; where the act is an argument and information is equal to explicans (premises) resulting in a conclusion or explanation of an event. This strikes one as being remarkably human in that we find ourselves in a world that demands explanation but in this very relation contains a necessary process of arguing for or against a number of causes – our success in this process is due to scientific causality.   

This notion is supported elsewhere in the text when Lewis expresses gratitude to David Velleman who told him that humans explain by way of analogy moving the unfamiliar towards the familiar. After discussing how its possible to explain in a bad or good way Lewis shows the shared interest we have in understanding by way of logical argumentation, ‘But credibility is not a separate merit alongside truth; rather, it is what we go for when seeking truth as best we can.(Lewis, 1986. 218)’ The idea that truth and credibility are to be taken on merit is then met with the capabilities of the human. The struggle to explain is just as important as the explanation itself and this is a big contributor to the power of science: it is an assumption to suggest that since our species first breath we have striven for the certainty the truth provides us because today some people desire to remain ignorant to the wonders that science may bring.

Philosopher Samir Okasha adds yet greater emphasis on the human component of science but just after discussing objects being ‘multiply realised’ at the physical level (how physical entities take different forms in the observable universe) he explains this notion of sciences incomplete reducibility by discussing the concept of a biological cell. But, this multiple realisability just deepens our need to understand explanatory arguments or argumentative explanations.

‘So the concept “cell” can not be defined in terms drawn from fundamental physics

There is no true statement of the form ‘x is a cell if and only if x is …’ where the

blank is filled by an expression taken from the language of microphysics.

(Okasha, 2016. 57)’

Okasha helps further one’s inquiry by allowing for an approach to the dilemmas at stake via way of language. Viewing the language of science is useful because it helps in honing in on the reasons for supporting our beginning statement and the following conclusion. Although you might say that adopting a position that views language as the main evidence in favour of explanations being arguments being invalid because it reduces the beauty inherent to the simplification that is necessary for scientific certainty, in the form of equations and formula for example. In other words one main disagreement is that arguments remain prone to linguistic uncertainty and ruin the simplicity inherent to science by adding unnecessary complexity by generalising separate instantiations of existence. This one believes is an interpretation that could be used to refute my positive conclusion. Viewed from Okasha’s discussion on the antagonism that philosophers and physicists debate that the laws physics builds upon with their assumed truth are not quite irreducible to a perfect description of physical phenomena. Another example of this conundrum can be observed in Bradford Skow’s paper on Physical Explanations of Mathematical Phenomena ( Skow. 2015).

However, although important and relevant such discussions move too far away from an everyday reality into the more abstract and formal discussion on apodictic qualities of physics and mathematics as such. To understand why these discourses should not be seen to effect our discussion on the co-dependency of logical structures within natural language (causal histories) and explanations (phenomenal events) then we should return to an idea mentioned by both Lewis and Okasha. The ‘covering law’ model first suggested by philosopher Carl Gustav P. Hempel states that if arguments are to provide causal information on an event then they need to appear in the form of a deductive nomological argument (containing only law premises and particular fact premises). The argument is deductive so if the premises are true then it is necessary that the conclusion is also true meeting the requirement of certainty science demands.

Yet, Lewis explains how Hempel also approached the different scenario of probability. Introducing a need to consider ‘the “specificity” of an act of explaining as being relative to the state of our knowledge; so that our ignorance can make correct an explanation that would be incorrect if we knew more.(Lewis, 1986. 232). This could bring in to doubt the belief in a human’s capacity to guarantee that there explanations can come in the form of an argument. But, there is one more contributory factor that I would argue supports an acceptance of the incompleteness of knowledge and that is the extremely relevant contemporary importance of information. Lewis also comments on information inviting us to consider its role in determining whether or not our explanations are of a good or bad quality – in fact information is the first on his list. I argue that this is structurally important for causal histories.

I have chosen to represent this by showing how the notion of a covering law argument also has to embrace a dualism or vulnerability seen as relevant to the physical state of our knowledge (information). In other words this necessary vulnerability in science pared with its certainty or inevitability of explanation are strong evidence for explanations being arguments. This is represented by two formal arguments below that show a certainty in explaining and then a vulnerability in whether our argument affirms or denies. I believe science needs its constructive dilemmas otherwise how would it continue to progress? One last consideration to further the scope of the essay is it important to avoid the trap of arguing for explanations under the guise of completed facts because these are always subject to change? So, subsequently it has to be the case that explanations should be arguments leaving the horizon of scientific discovery truly open to future human understanding[…] 


Argument one


  1. Explanations should be arguments using pre-given information.
  2. Scientists use a formal language (arguments that contain certain pre-given ) to explain a given phenomena. (1.2, explanans/explicands).

  1. Therefore it follows that there is new information produced

of the given phenomena needing explanation. (explanandum/explicandum).


Argument two


  1. If good information then an affirmation, and if bad information then a negation
  2. There is good information or bad information
  3. Therefore there is affirmation or negation








Lewis, D. (1986), ‘Causal Explanations’ in Philosophical Papers Vol. Ii. Oxford University Press.

Okasha, S. (2016), Philosophy of Science: Very Short Introduction, 2/e, Oxford University Press.

Skow, B. (2015), British Journal of the Philosophy of Science, 66, 69-93.  


The No Miracles Argument is a Decisive Refutation of Antirealism

At the heart of science resides sceptical or radical doubt; inherited from its founders. Figures like Descartes, Leibniz, Newton, Galileo all embodied movements of doubting what was perceived as real. Science has moved to a point where it appears trapped in a desire for absolute certainty in a set of physical laws affirmed in one equation and one mathematical proof (Hawking, Michio kaku, Thomas Nagel).

But, if we go back to the days of Descartes and Galileo there is a clear antagonism between free-thinking (doubting our understanding of the physical world) and the certainty of religious belief. Science has since seemed to be victorious in these disagreements. However, this success has come without physics being able to provide a completely certain explanation of the reality we exist in. This invites within science itself a physical reflection. Culminating in a contemporary debate involving those who support the idea that the conclusions science provides contain real facts that tell us something true about this world and its phenomena.

This stance is called ‘scientific realism’ and those in opposition  to such a perspective argue that science only provides “empirically adequate” descriptions of the unobservable phenomena; a position called ‘Ant-realism’.   The realists have used an argument called the ‘No Miracles Argument’ (N.M.A) to refute the Anti-realists. This argument supports what philosopher of science Hilary Putnam once expressed, ‘Realism is the only philosophy that does not make the success of science a miracle’, Putnam is supported by the vast evidence that science’s predictive force is highly successful (but, we should probably say reliable?).

Nevertheless, I believe the position of the realists and their use of the N.M.A does not provide a decisive refutation of Anti-realism. So, in this essay one will argue that a simplification of thinker Colin Howson’s thought on the N.M.A ; following Howson I propose, or put forward a position that expresses a simple model acceptance of miracles in science. In other words I think science does contain miracles suggesting that miracles can also be scientific. From this an argument against a realist use of N.M.A can be made: when realists reject miracles they also reject possibility and plurality in favour of necessity and singularity. I will now offer examples or contexts where evidence for this argument and conclusion can be observed.

Starting with the philosopher Colin Howson’s work on David Hume (Howson, 2015) we see how, ‘Hume inferred an extreme smallness of P(m) , from the definition of a miracle: as an event which violates the laws of nature.’(interesting “violates” the laws of nature)… It is possible to see Hume’s thought clearly: you don’t see a miracle everyday. But, this just remains trapped in observability which is too simple. Howson begins by showing that the N.M.A commits a ‘base rate’ fallacy, a fallacy that ignores or privileges one kind of information over another. He shows how the argument that supports N.M.A to be false it does not say anything about base rates or likelihood.

  1.  P(S/T) is quite large
  2.  P(S/¬T) is extremely small

  • Therefore, prob. (T/S) is large.

Where (t) is ‘substantially true’, and (s) is predictive success. This argument ignores the dependency of the posterior probability on the prior. That can be observed as necessary including likelihood (λ). Observing odds can be seen in ‘Bayes Theorem: odds (T/S)  = λ odds (t). Where odds are related to probabilities in the usual way, and (λ) is the likelihood ratio, so P(S/T)/P(S/¬T) , this then only shows Bayes factor in favour of P(t), and that likelihood is large; nothing about the or its odds. Thus being fallacious because as Howson points out P(t) does not have to be very large to generate a high probability value.


In contrast to this a separate thinker named Psillos who attempted to reformulate the N.M.A so that it would acknowledge or consider evidence (Psillos 2009).

  1. f =1
  2. f =< 1
  • f =0
  1. f is close to 0
  2. S is the case     /  Therefore, impact of S on P (T/S) > P ( ¬T/S)

I agree with Howson’s rejection of Psillos attempt to support and re-articulate the N.M.A. In short by referencing the fact that probability coherence needs consistency. Instead of being able to choose or hand pick agreement between (t) and a given observation Psillos shows that success tells more in favour of truth than falsity because what tells in favour of truth depends on the prior.

In response to these two related examples I would argue that the N.M.A can not be seen to refute Anti-realism: that is,

(rejecting miracles simultaneously rejects possibility and plurality in favour of necessity and singularity: ( ¬  ◊ (p) → □ (s)).)


Phlogiston Theory is simply false, because phlogiston does not exist, and has been entirely superseded by the theory of oxygen.

Let us question this statement and see what it can communicate. First, in this statement we see a negation the claim is that Phlogiston theory is “simply false”, and then two explanations: 1) it does not exist, and 2) it has been surpassed by the theory of oxygen. So, our question may initially be twofold does either the discovery and theorisation of oxygen by French scientist Lavoisier make Phlogiston theory false and does the fact that phlogiston does not exist today make the theory worthy of simple falsity? In our discussion it will be greatly beneficial if these questions could be asked in a way that clarifies both the truth and meaning of the above statement, its position, and relevancy to the wider practice of the philosophy of science.

We could begin by suggesting or adopting the most popular definition of Truth still used by science today. That Truth is one, and a continuation of this one (it holds true and remains true over a period of time: 1-1-1-1-1-1-1 …). Is it acceptable to suggest that the success of a theory is completely dependent on its truth preserving abilities? Here it would appear that if we take a science as a whole we see the legacy of the ancient Greeks Parmenides and Aristotle. The ‘Principle of Non-Contradiction’ in Aristotle is so influential it states that it is impossible for one thing to be true and false in the same way and at the same time. This could be easily taken as sufficient to affirm that indeed Phlogiston theory is simply false because oxygen clearly serves as a better explanation for a substance that when released enables bodies to burn.

But, let us look at this in an argument form:

  1. Phlogiston does not exist
  2. Phlogiston theory has been superseded by oxygen

  • Phlogiston theory is simply false

I see two discrepancies with the initial premises they do not appear to directly lead to a simple falsity. In that the first premise claims phlogiston as a material thing does not exist. The second premise says phlogiston as a theory has been transcended and replaced by oxygen. The problem here is somewhat obvious: 1) to what extent does a theory hold true to a reality subject to change? Or, do phlogiston and oxygen refer to the same object? Then, if we look at the conclusion we have assess the falsity of the phlogiston theory. If we were being very critical we could also add that the use of an adverb to describe falsity invites in an auxiliary line of questioning. However, let us keep “simply false” as meaning the simple definition that it is not true.

To conclude:Today, I will fall In line with Thomas Kuhn’s thought in that the discovery of oxygen represents a separate paradigm and therefore the frame of reference (the capacity for technical terms within a theory to correspond) is also cut. So, they are two separate objects (almost as if they exist in different worlds or universes of discourse). This though is not sufficient to say it is “simply false” rather a better conclusion would state it is necessarily false in this world and at this moment. Thus, expressing a complexity of negation essential to understanding the truth value or true value of oxygen.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s