with No Comments

Post No.: 0473process

 

Furrywisepuppy says:

 

The reason why the scientific process is the most reliable way to obtain the truth is because of the testing and challenging of beliefs and hypotheses, and the need for evidence. Things aren’t just guessed at through intuition or faith – they’re empirically observed and measured. It’s arrogant to think that personal belief or faith is all we need to make something true or happen. We need evidence and this takes hard work in the form of research.

 

Our intuitions are fallible for we can perceive patterns in the world that aren’t really there thus it’s better to use the scientific process to test and observe if there (likely) truly are any patterns. Science is about scepticism (rather than cynicism), about empiricism, and about lots of repeatable experiments. The scientific method isn’t so much used to prove that we’re right – it’s used to risk proving that we may be wrong.

 

The scientific method aims to improve objectivity and thus eliminate human biases as much as possible – it can achieve this via experiments that make use of blinding, by using a large enough number of samples or participants that or who are objectively randomised yet representative so that any conclusions derived from them can be confidently generalised to the wider picture or population, by using control groups to make direct comparisons with, by systematically manipulating (preferably) only one variable at a time, and by carefully recording everything because our memories are unreliable.

 

There’s the use of statistical techniques, data analysis, contingency tables and comparing effects to chance (‘rejecting the null hypothesis’). These aim to quantify the information. So when some research claims that some outcome is ‘significant’ – this can be quantified as a number that’s objectively greater than a certain critical probability (usually, we want to be at least 95% certain that an effect is not due to error or chance), for instance.

 

There’s also the questioning of assumptions, the application of critical thinking, and the publishing of information so that other people can scrutinise and challenge what we’ve done (the independent peer review process) or try to replicate the results of a study or experiment (replication studies) for themselves. This can mean that scientifically obtained conclusions might change, which allows them to refine, improve or even totally revolutionise over time.

 

This isn’t a fuzzy criticism of science but the way that knowledge grows and refines – the scientific community adjusts its views based on the evidence, which can mean as a result of discovering new evidence (e.g. due to progressive advances in the accuracy of the tools to measure phenomena, or due to having access to more data as more and more people in a population get tested for a disease). We don’t have access to everything in one go that we could possibly want to find (e.g. in the past, people from the northern hemisphere didn’t know that black swans existed until they travelled to the southern hemisphere). Science is a messy process full of successes and failures, potential changes and potential changes again. And it’s ultimately better for it because this is better than not updating one’s beliefs despite what the aggregate evidence, or disputed evidence, shows.

 

Bias can however still creep in when presenting the data (e.g. cherry-picking data due to conflicts of interest or outright research fraud) or when interpreting the data (e.g. the significance or qualitative meaning of a result) – hence one is best off examining entire datasets of results and not just the select few data points or summary statistics presented by the media headlines, which often just lazily or oversimply parrot the author’s own given conclusions. Scientists are also human after all and thus subject to personal biases, motivations and genuine mistakes, slips and lapses. These are where errors in the scientific process can come in – and these can be deliberate (motivated fraud) or accidental, and done consciously or subconsciously. This is where the values of the peer review process and replication studies come in.

 

Understand that science doesn’t pretend or aim to answer moral/ethical ‘ought’ or existential ‘meaning’ questions – only ‘is’ and ‘how’ questions. Although science can usually usefully help guide the former, science cannot help answer all socio-political or moral/ethical questions. Science itself has no moral compass (science itself needs to be regulated against unethical experiments e.g. it’d be a poor defence to claim that injecting babies with heroin to see what happens is ‘just science’(!)) So science may reveal some facts, but what we decide to do about them is a socio-political (i.e. democratic, philosophical, debating) process – unless we want a scientocratic society (which is a form of politics anyway). For instance, science clearly and consistently demonstrates that surveillance increases cooperative behaviours – people are less likely to cheat or steal when (they feel like) they’re being watched – but should we therefore fully embrace surveillance wherever cooperation is beneficial to societies?

 

Scientists therefore shouldn’t be conceited in thinking they have all of the answers; and even when they do have answers for some things, they shouldn’t act as if they don’t need to consult anybody else. Populism can arise as a counter-reaction if those who are considered part of the elite (in this case the intellectual elite) treat those who don’t understand as if too stupid to listen to them. The skill of communicating science to others is a social intelligence skill, and being patronising is not socially intelligent and therefore not persuasive in making scientific facts socially accepted facts. We certainly need to involve citizens in ethical decisions posed by science, such as regarding the collection of personal data and what to do with it, even if it’s for some greater good.

 

Gun controls might objectively reduce crime according to the data – but there are trade-offs without an objective answer, such as does a polity value its freedom over its own lives? This sounds absurd but some things are perceived as worse than death, and science cannot answer whether freedom (in the form of private gun ownership at least) is valued more highly than people’s own lives. Well actually, the answer comes in the form of a self-reported opinion – which is in essence a political vote. (We could therefore possibly argue that a political election or referendum is a form of scientific study on a particular population? The debates will then come in the form of whether we think it’s barking mad that some people value their freedom to own guns over their own family’s lives, how well-informed we think each other’s views are (such as someone who values his/her own family members’ lives above the freedom to own guns yet erroneously believes that owning guns is how to protect his/her own family), that ‘gun controls’ come in many different forms and we haven’t scientifically studied them all yet, etc.…)

 

Regarding abortion, it’s arbitrary or subjective to state that the definition of ‘alive’ is the starting point of consciousness. Or even if we all agreed with this – how shall we determine this in practice (e.g. shall we use independent movement, an independent pulse, a defined face or a defined heart)? The scientific community cannot even unanimously agree with whether viruses are dead or alive because ‘alive’ can be operationalised in many different, subjective, ways. The same kind of questions can be posed for artificial robots or machines and if they can ever be considered alive (since every living organism is essentially a machine too)? Voluntary euthanasia is probably another perpetual moral/ethical issue that cannot be solved just via science. Read Post No.: 0461 to examine the issue of operationalisation more deeply.

 

So don’t assume that what science or scientists say is always objective – in many cases, there are still personal human choices of choosing definitions, such as what measure to use for ‘aliveness’, ‘fitness’ or ‘intelligence’? Does ‘problem-solving ability’ mean spatial, financial, relationship or some other kind of problem? Where the line is drawn between someone being classed as ‘pre-diabetic’ or ‘diabetic’, or ‘overweight’ or ‘obese’, for example, is ultimately arbitrary because there’s usually a smooth continuum between ‘low health risk’ to ‘high health risk’ rather than clear tipping points, hence how much is ‘too much’ or ‘too little’ of something? Is a community where no one wears any shoes – on the face of it – a fantastic market for a shoe seller or a terrible market for one? This all comes down to personal human interpretations and predictions.

 

We therefore sometimes lack objective ways to be objective! This is why we must always read the methodology of an experiment or study, then its results, and read up on other experiments and studies conducted in the same area too, then come to our own conclusions rather than take a scientist or journalist’s conclusions at face value – science is the process, not a book of dogmas or some other kind of set-in-stone bible. Even when the scientific process is well-conducted, some things are by nature more disputable than others, even though they may all fall under the umbrella of ‘science’. Different scientists sometimes strongly disagree with others in areas that are inconclusive.

 

…Yet if we do dispute a scientific study’s findings or conclusions – we better be sure we can present some scientific research to support our own conclusions! This might mean pointing out some other research that has fewer methodological flaws, or carrying out an experiment of our own and publishing that for peer review. Otherwise we must be fair and go with that given finding or conclusion as the ‘the best truth we can say right now’. The sceptical mindset is about being able to go from one ‘best truth we can say right now’ to potentially another. We must therefore never stop learning and keeping abreast with the latest state of the aggregated pool of overall knowledge gathered so far. We cannot do better than this because we can never know for sure whether we’ve learnt 100% about something. (Metaphorically, or perhaps even literally, it’s like possibly later discovering that there are green or red swans living on another planet. It may be an infinitesimal chance but it’s not an absolute zero chance.) That’s why all scientific conclusions should really be regarded as provisional in case we discover something in the future that contradicts everything we thought we knew. We haven’t discovered and understood everything in the universe (and beyond?) yet, and we’ll never know if we have even if we one day will.

 

The utility of science is in making generalisable conclusions, extrapolations and predictions about the future and/or wider picture, and these can never be absolutely 100% certain. History is the subject for merely recording historical events, and we can potentially be absolutely 100% sure of what happened in the past because known history is already done and cannot be changed.

 

So, many conclusions from manuscripts or papers that have been peer-reviewed and published by reputable journals will be pretty reliable, but some will still be less certain than others. Laypeople tend to read scientifically-derived conclusions as black-or-white but they all fall under various levels of confidence. A minimum of ‘two-sigma’ (95.45%), ‘three-sigma’ (99.73%) or ‘five-sigma’ (99.99%) is usually chosen depending on the context, but that means there can be as high as a 1-in-21.98 risk of a result being down to error or chance. Perhaps all reports of science news should therefore point out the confidence level of a result?

 

Although faith seems to be exclusive to religious, superstitious or irrational beliefs, there are some things – fundamental things – in science and mathematics that must be taken as faith, as in ‘it just does’ or ‘it just is’, too (e.g. quantum superposition and entanglement, or Gödel’s incompleteness theorems showing us that some mathematical axioms must be accepted without proof because they logically cannot be proven).

 

But where things can be proven, contradicted or discovered – we must seek those proofs and discover the evidence via the process of science; and then heed them.

 

Woof!

 

Comment on this post by replying to this tweet:

 

Share this post