THE NECESSITY OF DATA

Have you ever waited on someone for what you thought were hours till he reminded you that you have only been waiting for fifteen minutes ?

This is why having a way of checking what you believe is important. We can be fooled into unnecessary exaggeration, and in other times underestimation. I propose that the best way to test a belief is against reality. That’s how it can be reasonable. I take it for granted that reason should be followed by impeccable logic. It helps when an idea is internally consistent. But this isn’t enough for it to be reasonable. Imagining things in your head can have great internal consistency, but these imagined thoughts will be unreasonable if they don’t map to facts on the ground.

Take an example of a laudable ideal like spreading democracy. And doing this through violence. Will it be reasonable to change regimes when it leads to so much bloodshed ? I will agree that every ideal requires some sacrifice. But is the sacrifice worthy of the price ? Other pertinent questions will include : Are those living under these dictatorships complaining ? Will the future ‘democratic’ regimes make the people worse off than the status quo. I intentionally put democratic in inverted commas because opponents to a regime are not necessarily democratic, and can pretend to be democratic in order to win power especially if they see the possibility to be backed by powerful Western Nations. I believe democracy is the best system of governance yet. What I don’t believe is spreading democracy no matter the cost. I often hear people say some things are to be done no matter the cost. One example was when the president of Ghana, Nana Akuffo Addo, said after starting his flagship Free Senior High School Education (Free SHS). He said it didn’t matter what the cost is. This kind of thinking is untenable. Especially a socio-economic decision. We could say what he says is political talk. That he doesn’t mean it. My main problem with the policy is that there is no option for those who can pay. I realized in writing this essay that I needed to submit myself to the principle I am discussing. I tend to dislike policies that appear to be socialist ; although I have benefited from the government’s National Health Insurance Scheme (NHIS). In Ghana, it makes no sense not to have the NHIS card because a single visit to a hospital without the card will cost you more than the yearly premium that the scheme requires. It hit me that being anti-socialist isn’t enough. I have actually changed my mind on welfare. I think some kind of welfare might be necessary. My change of mind was influenced by reading Karl Popper’s Open Society and Its Enemies. In relation to welfare, he states that we are not responsible for the happiness of others, or to do unconditional good for them. Our duty as humanbeings is to save others from distress. I am still paraphrasing Popper. And the role of government can be to fulfill this role so the temporary poor don’t wallow in suffering. So that the rich don’t enslave the poor, as the poor have something to live on as they plot their comeback. This is different from government doing good. The role of the state, and especially its laws should be to prevent evil. Not to do good. It is easier to know what is bad than what is good. This is why I dislike ideas of an utopia. It goes back to my problem with apriori thinking without contact with the ground which I expressed through the idea of spreading democracy no matter what. If spreading democracy leads to modern day slavery in Libya, and trying to create an utopia lead to millions of death among the Soviets and the Chinese, you don’t argue the point away. You don’t say for example that the catastrophes under communist regimes were not socialism properly practiced, as communism and socialism are not different in essence. The question is what are you willing to sacrifice to see heaven ? The answer is often everything. I don’t like that answer because in it no sacrifice is enough. I also think that the idea of a socialist paradise is naive. It is an ill conception of human nature. Although those of us socialists will decry free markets, they tend to think that these same humanbeings who run the extended order will change and become saints in a heaven on earth, where there is no suffering. I don’t believe that. I don’t also buy the argument that humanbeings are inherently good. People are good. People are evil. People are people. The vision of an utopia misunderstands human behaviour at scale.

My mum keeps saying that people are dying younger than they were in the past. She has what she thinks are the causes of these early deaths. I had the impression that she could be wrong, and I keep telling her so. I remind her that there are more people alive now than in the past ; therefore she is likely to see more younger deaths. This is to say that she is tricked into her position by the availability bias. This is a point against her position, but it is not enough to prove that she is wrong. To do that I need data. In preparation for this essay I found a graph that plotted life expectancy over time in Ghana from the World Bank Website. This showed that life expectancy in Ghana has been increasing. I am no statistician. I feel my mum is likely to be wrong, because for nothing at all we have access to drugs and cheaper food. In the spirit of the title of this essay my feeling and even the data point isn’t enough to disprove my mother’s position- it isn’t enough to discredit her position that people die younger now. For her, the youth are abusing substances and involved in lots of moral depravity. Why do I venture to say this given that the World Bank Data supports my position ? Life expectancy is influenced by infant mortality. Now that infant mortality has been drastically reduced, it follows that life expectancy will increase. A better analysis of the problem will be to look at the life expectancy of those people who reach certain ages. An example will be this- how has the life expectancy of five year olds varied over time ? This sort of analysis will give a better answer to the problem. Alas, I am no statistician. I may not know how to resolve this simple problem.

I tried to explain how problems can’t simply be solved with data. This isn’t to say they can’t ultimately be solved with the help of data. To be more specific, answers to questions that plague us can be solved by looking at data. However, the data point should be analysed with the question posed in mind.

I was motivated to write this essay after reading Enlightenment Now by Steven Pinker. It is a great book that illuminates enlightenment values. One of these values is knowledge. The logical claim is that knowledge should be reality based. Thus, utterances shouldn’t be supported by mere opinion, but facts. In this way we avoid the availability bias I have already mentioned.

However, there is an undercurrent I find uncomfortable. Pinker seems to confound different classes of risks. I am being careful. It is for this reason I say he seems. I am no expert. He dismisses fears about nuclear energy by using data from catastrophes that happened in nuclear plants in the past. I am not sure if he qualified what he wrote. He discusses black swans in the book, but he didn’t seem to apply it to nuclear energy. A black swan is an unpredictable event with great consequences. It highlights the problem of the inductive method in science. No matter how many white swans you see, you can’t make a logical conclusion that all swans are white. This leads the great Karl Popper to prefer falsification. Popper’s method, if I understand it well, is to get science closer to deductive reasoning. So you start with a theory or hypothesis, and you carry out experiments to falsify it. Ultimately, the goal of science is not to say something is true. Rather, that it is not false. There is also the practical aspect where a theory can be true enough. This is seen with drugs. They don’t have to cure every single diseased person for them to be useful. The black swan idea was made popular by Nicholas Nassim Taleb in his book of the same name. Taleb is a probabilist.

Pinker’s apparent confusion with risk categories led him to a confrontation with Taleb. This was after the publication of Pinker’s book, Better Angels of Our Nature. I haven’t read that book. Maybe I will pick it up later. Till then, I wasn’t familiar with Steven Pinker. I am a fan of Nassim Taleb. I have read all his nonmathematical books. Taleb is a genius. As we are all not perfect, sometimes his opinions are too strong. In the confrontation, Taleb gave the impression that Pinker was an idiot. I read a little about Pinker on the internet. In the little reading I did, Pinker said it wasn’t necessary to do too much school, and that it was usually better to start a business. I thought someone with such an opinion couldn’t be too bad. Taleb first challenges the figures in Better Angels of Our Nature saying, for example, that war parties are motivated to exaggerate or underreport casualties. Taleb’s second argument is that what’s more important is possible magnitude of harm, not actual crime. And with nuclear weapons the possibility that the entire human race can be wiped out is not zero. In Enlightenment Now, some of Pinker’s analysis reflects an inability to understand risk. I am no expert. But when he says things like nuclear weapons haven’t killed many people, as compared to traditional ammunition, and that even Hiroshima has been rebuilt, it seems to me that he just doesn’t get it. This is not to say that his central message in the two books are not necessary. His central message is that things have got better, and they have been getting better for a long time. However, you wouldn’t know this by watching the news. Gratefulness isn’t a bad habit to cultivate. Besides, if things are getting better, and we know how and why they got better, we can accelerate progress. The problem with Pinker is his underappreciation of risks with black swan properties. His way of thinking will lead him to say that there is no reason to worry about Corona Virus deaths because more people die from road accidents. He would be led by his logic to say something like this at the onset of the virus. I am only guessing what he might say. It can be safely predicted that road accidents will not increase by even 100% in a year. But for a virus there is simply no cap. A viral infection can become a black swan event.

It is important to appreciate risk categories in order to avoid the Turkey Problem. The Turkey Problem is a story about a turkey who gets well fed. In its mind its owners love it, that’s why they are fattening it. Unbeknownst to it, it’s been fattened for Christmas. Sometimes data doesn’t give you good information. It can actually mislead you. Think about a gambler who recklessly increases his stakes but keeps winning. He would be increasing his ultimate ruin. This isn’t to say data isn’t important. Rather, it is necessary for us to adopt a critical attitude in its analysis.

It seems to me that somethings are worth fighting for no matter the data against it. This is especially so when you have nothing to lose. For example, you have nothing to lose if someone wants to make you a slave. I think I am wrong. I remember Frederick Douglass saying in an autobiography that the main reason why most slaves found it difficult to run away was that they wouldn’t be close to their families. So there can be compromise even in the face of slavery. What I am getting at is that some things are worth doing even if the data doesn’t support them. I am finding it difficult to clarify this point. I think this leads back to what I said about data being a guide to action. Data is guide, not substitute. If something didn’t work in the past, data can help us identify why they didn’t work, and might help us find ways to increase our success. There is the tendency of overwhelming data against an action lead to a defeatist attitude. I am not saying we should push forward despite overwhelming evidence against the action we wish to take. Like those who hunger for an utopia despite the overwhelming evidence that such trials lead to overwhelming suffering, and that free markets with its accompanying private property works fine despite its problems. When evidence is against us we should take small bets. The situation that usually presents itself is that we have to commit to an action when we are unsure of its consequences- like starting a business. The startup example will help me clarify what I want to say. It is said that most businesses fail. I forget the exact fraction. It is high. I am not sure how true that assertion of failure is. Josh Kaufman who wrote Personal MBA explains that there is a problem in the methodology that accounts for the failure because businesses who merge with others are classified as failed. It’s the same with registered businesses whose owners don’t file tax returns. It isn’t obvious that a business which is not working has failed because the owners might have learnt their lessons and started a different one. Let’s now assume that the assertion that most startups fail is correct. So what do you do about it ? Do you fold your arms and do nothing ? I agree that it is foolish to try to move a mountain with old tools that didn’t work. However, a new business is not every dead business. Being defeatist is not a useful strategy. This is where I think faith is useful. Faith isn’t blind belief, or atleast it doesn’t have to be. I am using the word faith in a sense that is devoid of religious connotations. Later I would try to explain why I think religious faith is nested in logic. Going back to having faith in action. It’s the problem of whether the glass is half full or half empty. This has nothing to do with facts ; it depends on ones interpretation. Is life meaningless or meaningful ? Which answer is more useful ? I hear some people say that we are a tiny speck in the universe, which often leads to the logical conclusion that nothing we do matters. Or that we will eventually die together with all those we love, so life is a net loss. So what are you going to do ? Sit there depressed and do nothing because all is vanity. This is a useless way of thinking. You can’t be grateful with this mindset. Not everyone won the lottery to live to experience consciousness. And there is so much good to be done in the world.

I have a problem with the popular conception that beliefs can’t be justified. I am no philosopher. I might lack the academic erudition to attack this idea, but I will do it anyway. There are two senses in which beliefs are often said to be impossible to be justified. The first is that we start out having beliefs without having evidence for them- it’s this critique I intend to argue against. The second is that what can be analysed are actions, not beliefs. People can lie about their beliefs, but actions are for all to see. But this also leads to a practical problem. Like Apostle Paul in Romans 7, we do things we don’t want to do, and things we want to do we don’t do. So action isn’t enough to say what someone believes in, although it can be a good predictor. Another practical matter is that professed beliefs do not matter as much as the actual behaviour of its adherents on society. However, the first sense is invoked when people say that beliefs can’t be justified. I earlier spoke about the usefulness of faith. I also tried to establish that faith is not equal to stupidity. Even choosing to follow reason is an act of faith as the great Karl Popper points out in Open Society and Its Enemies. This is not to say it is an arbitrary virtue. If you will not choose reason, what would you rather choose ? Choosing reason is an act of faith because we don’t know before hand all the consequences of choosing reason over, say, superstition. I am uncomfortable with this contrast because I think it can be reasonable to choose a superstition over uncritical rationality. So when I contrasted reason with superstition, I meant harmful superstition. As I have said, beliefs don’t have to be arbitrary. We see people act. We realize from our lived experience, and observing others that being reasonable makes us flourish. There is also the fact that truth often maps on to reality. When I venture to say that beliefs can be justified, I mean that they are adopted based on observation. Although we are not privy to all the results of human actions, we often have an inkling of how to get the best results possible. This is a belief. Also having such a belief can be independent of whether we are able to act in alignment with them. However this leads us to the problem of induction. But the problem is of little importance. The problem of induction is especially of little importance in the game of life. Honesty is the best strategy doesn’t mean that you will win every single time you are honest. It can lead to your death. However in the long run it can lead to success. It allows you to play a repeated game with others, as others allow you to play with them because you play fair. The best team in a league doesn’t have to win every single game, it has to win the league. A good belief doesn’t make you win in every single situation, you live a good life.

This is the very reason why I mentioned earlier that religious beliefs are nested in logic. You see people act, and you try to emulate attributes you think will lead to your flourishing. You might accept that some attributes would lead to blossom even if they are difficult to practise. There is also the problem that we often don’t choose our beliefs. Where we are born plays a more influential role than we care to admit. Imitation of an attribute doesn’t necessarily mean it is useful. However an attribute that doesn’t map on to reality will go extinct eventually. And despite the above problems I have mentioned- how the environment affects our beliefs and the uncertain utility of imitation, I hold that we adopt beliefs by watching others act, and ideally choosing the most profitable path. Again, we don’t have to worry too much about imitation which is useless, read harmful. Those who practice them will vanish in the evolutionary timescale. Actually, the useless isn’t a problem, just the harmful. The useless can have a use we don’t know of. The harmful is just what it is, harmful. Of course religious beliefs like my earlier example of faith without ornaments falls prey to the problem of induction. So what are you going to do ? Lock yourself up in a cave ? Flee from humanity ? A belief can be true enough.

Lest we forget the central subject of this essay, I will repeat it- data as guide to action. A good guide is one that has helped its adherents survive. A truth in human interactions is one that helps its adherents to survive as they play a reiterated game. People often confuse moral truth claims with scientific ones. There can be intersections, but they often deal with different realms. Moral or religious truth claims are valid based on their effect on human action. Scientific truths almost never tell you how to deal with other humanbeings. Religious truth claims are often useful fictions. Their useless elements those of us who are scientistic attack are often irrelevant. They end up throwing the baby out with the bath water. Existing belief systems are better seen as having survived in a pool of other beliefs. Imagine all these beliefs fighting for survival. Like natural selection, groups are selected by the environment, or how well their beliefs map on to reality. Group selection in human action is different from biological selection where there is consensus that selection is at the level of genes. It seems to me that what’s more important in religious beliefs is how it makes people behave. If it makes people behave in a way that encourages human flourishing, then it’s good.

This essay is getting longer than I planned. I will like to end by adding this information I have found very useful. It is to use data in personal life. If you are acquiring a skill, time yourself. Count the number of push ups you do. The number of minutes or miles you run. Maybe how many pages you read each day, and the titles of books you read. These habits have helped me overcome exaggeration and underestimation. Like the waiting time example in the introduction, if you have a watch, if you have data, it is not easy to lie to yourself.

Thank you for reading.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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