I have the feeling that theoretical science seems to be dead and needs a rebirth. Data science is in; where you have people studying patterns and trends in data, and a lot of what scientists are asked to do is fix the data or come up with new ways to study data.

Observation and logical inference are out. Data collection and analysis are in. The things is so many tools have been invented to study data, including infrared cameras and huge data centers that study gazillions of data points which make data collection and analysis a lot easier. So participant observation or data observation are kind of outdated methods of science, or of little interest to those who think any scientific claim should be backed up by data.

The problem with data is that it's hard to study the cause and effect relationship in any data set that is what caused a phenomenon to be produced, unless there's other data to help study the cause and effect relationship. But with observation you can infer the cause and effect relationship.  

But data science misunderstands the true nature of science. Science is supposed to help individuals take decisions in the concrete world, be they new policies or new ways of building physical objects. To me, observation works just as fine as data when it comes to taking decisions. Actually, with observation, a scientist is in a position a lot similar to that of a politician or an engineer in that the politician or engineer can make the same observations.

Let's look at it this way. Observation is field work that is you are watching nature at work, observing and describing. That puts you in a position to see nature alive, and that makes you familiar with nature, as you “speak” nature's language so to speak.

Data on the other hand is computers observing nature at work. Computer collects data, analyzes data and gives you the observations. Problem is: if you're not familiar with the theory and how nature actually works, you won't “speak” nature's language and you won't be able to understand that data analysis and its conclusions.

Kids these days want to skip theory and observational studies and want to move straight to data analysis, because it's easier, and they get to stay behind their computers, which is an area familiar to them. Problem is, while they are good at trusting their computers and the data provided, they tend not to understand the phenomenon that they are observing.

Let's look at it this way. Most of my theoretical work was observation, as I was a “student” of the phenomena that I studied before describing them. Be it the work I did in languages, astrophysics or the economy, I lived through the phenomena. I was unemployed for many years (and still am) which is why I understand unemployment better than the average person. I also worked indirectly with banks and with the financial system which helps me get a better grasp of economic phenomena. I also spent hours contemplating the universe. And of course I learned several languages which helps me understand how language works. And I infiltrated several social groups which helps me understand how they work.

But if you're going to input data on how language works when you never really studied a foreign language, you could have trouble understanding the implications of the data study. How are you going to try to influence a policy-maker on language when you yourself don't really understand how language works. It's going to be hard to convince any policy-maker of the relevance of your study.

But, popular interest is vital to scientists. It is the air we breathe. These days, people are obsessed with COVID and with self-isolation and confinement and masks and social distancing, which makes a lot of the scientific work that we do irrelevant.

How do you elicit interest from a reader on any scientific topic when the reader is trying to figure out whether vaccinating against COVID really protects from COVID and has no other real interest. You're going to have a rough time convincing readers of the relevance of your scientific work.

It gets even worse than that. At one point, as a scientist, I was asked to stop writing because my scientific research deviated the reader's attention from COVID issues and the risk was that they would read my paper and forget COVID is even around. That's how silly things can get.

So indeed scientific research is done with the goal to elicit policy makers and other individuals to take decisions, and to try to influence their decisions on what is scientifically-backed information that voting such policy can work or designing things a certain way can work. But if policy-makers are too busy fighting COVID, you can't really catch their attention.