Simplicity in science
- The scientific method is the best way we have for solving problems and coming up with explanations.
- Science is self-correcting and always improving.
- As we learn more, explanations often become simpler.
Here's an example: When it was thought that Earth was the centre of the Universe, all sorts of complicated mechanisms and mathematical models had to be developed in order to explain the relative motions of stars and planets. When it became clear that Earth actually moved around the Sun, the mathematical models were simplified greatly.
- Ockham's razor: developed by William of Ockham (or Occam) who lived from c.1285-1349. He wrote 'Don’t make more assumptions than the minimum needed’.
- It really means ‘The simplest explanation is most likely to be correct’ or ‘keep it simple, stupid’.
- The complex answer may be correct, but always explore the simpler explanations first.
7.2 Scientific thinking: empiricism, scepticism and rationalism.
- Common elements that all scientists would recognise:
- Empirical evidence is something that can be observed or measured, and is usually quantitative.
- Empirical evidence is also repeatable, so data can be checked directly.
- Scientific evidence is not based on hearsay, testimonial or circumstantial evidence.
- It is the data that are important, not the fact an expert witness, or some authority or teacher produces them.
- The development of the idea of empiricism that formed the foundation for the development of modern science.
- Science is institutionalised skepticism: never believe anything until you see the data.
- Scepticism is different from cynicism, which is always looking to undo an argument and is negative.
- It is also different from denialism and contraryism, which are not open to new data or ideas.
- For knowledge to advance, it is important to challenge traditional explanations.
Rationalism: reasoning in science
The scientific explanations
that we come up with are the result of reasoning and logic.
- What is the premise of the argument?What assumptions have been made in establishing this argument?
- How do the arguments relate to the available evidence?
- Is it argument by example?
- Are there any exceptions, and are those exceptions important?
- Is the conclusion based on a sequence of thought or are there ‘leaps of logic’?
7.3 Inductive and deductive reasoning
- Inductive reasoning: based on observations of the world and can be used to form universal laws; it proceeds from the particular to the general.
- Deductive reasoning: based on universal laws and can be used to explain individual observations; it proceeds from the general to the particular.
- Figure 4.3: The relationship between inductive and deductive reasoning in science.
- Inductive and deductive reasoning can be used together to form explanations. Inductive reasoning applied to facts acquired by observation can lead to the formulation of theories and laws.
- Theories and laws can be used to form predictions (or to explain particular events) using deductive reasoning
Logical reasoning is essential, but in science, each step needs to be backed up with empirical evidence
- Aristotle (384– 322 BCE) formalised the rules for deductive reasoning as follows.
- Any argument can be reduced to two premises and a conclusion.
- “Given that (a) AND (b)…THEN (c) conclusion”
- You deduce the conclusion from the premises, which can be either a fact or an assumption.
- A valid argument is one where the all the parts (a, b, c) are present.
- However, it could be based on a false premise.
- A sound argument is one that is both valid, and correct.
- Role of experimental science to ensure that the premises are correct
Example of deductive reasoning
- An example of deductive reasoning:
- (a) All objects are drawn toward Earth via gravity (a variation on ‘what goes up must come down’). (Premise)
- (b) A ball is an object. (Premise)Then (c) If I throw this particular ball, it will be drawn to Earth via gravity. (Conclusion)
- Another example of deductive reasoning:(a) All grazing herbivores have shearing crests on their teeth which exhibit scratches due to wear caused by the silica in grass.
- (b) Teeth of Chaeropus ecaudatus have high shearing crests and wear scratches.
- (c) Chaeropus ecaudatus must have been a grazing herbivore.
Example of valid deductive argument that is not sound
- (a) All insects have wings. (Premise)(a) Slaters are insects. (Premise) [FALSE – slaters are not insect]
- (c) Therefore slaters have wings (Conclusion)
You can see that having a false premise can lead you to the wrong conclusion.
- Works back from the conclusion to the explanation.
- In science, this would mean collecting a lot of data and then coming up with an explanation for the observations.
- This is actually how a lot of science is done. Only later is deductive reasoning used to sort it all out and mesh the new findings with existing theories in the scientific literature.
- Inductive reasoning leaves us with a broad statement (which is often easily falsifiable - though that does not make it wrong!).
- To be ‘falsifiable’ means that a statement is testable by experimentation and that such experiments could yield results that show the statement is false.
Example of inductive reasoning
I make the observation that each time I get up before 6.00 am, I am extremely tired by 4.00 pm in the afternoon. I reason that getting up early makes me tired.
- Another example of inductive reasoning:
- Many, many observations are made of the teeth of mammals that eat grass.
All of the observations indicate that such teeth have high shearing crests that exhibit wear scratches.
The conclusion (based on inductive reasoning) is that all grazing mammals have teeth with high shearing crests and wear scratches.
7.4 Popper and Kuhn on Logic
and Thomas Kuhn
were both 20th Century Philosophers of Science.
- Popper argued that the correct way to implement science was the hypothetic-deduction method (i.e. starting at step 2 in the scientific method cycle described in Chapter 3).
- Thought that scientists should start with a hypothesis, work out what to expect by deduction, perform tests and compare the results with the expected outcome.
- Kuhn observed what scientists actually did, rather than philosophise about how science should work.
- Found that most scientists actually started with some data and THEN make a hypothesis using inductive reasoning (start at step 3 in the scientific method cycle and then go back to step 2).
- In practice what really happens is a bit of both.
- Often you follow up your initial idea by making some observations or doing a few small experiments, then, based on the pilot study, you form a hypothesis that you test using deductive reasoning.
7.5 Falsification versus rejection
- According to Popper, scientists should try to disprove hypotheses rather than prove them.
- This concept was formalised by Popper as falsification (can be falsified).
- it means that it must be possible to conceive evidence that would prove the claim wrong, or that a negative result is possible.
- Finding something wrong with a hypothesis is always definitive.
- If, however, your data support the hypothesis, then it might be true, but it could also just be a coincidence.
It remains a critical part of scientific thinking that a hypothesis is only scientific if it can, in principle, be proved false.
Why do some people reject scientific evidence?
Humans like consistency and there is strong tendency to eliminate cognitive dissonance
: if the theory or data does not agree with your ideas or philosophy, then it is common to reject the science.
- Examples include:
- the anti-vaccine lobby climate change denialists.