About 4 months ago I switched into “The Social Sciences” and got completely overwhelmed. I studied Computer Science at ETH Zurich before and anticipated that it would be quite different to study something like “Governance, Economics and Development” at a Liberal Arts and Sciences College (Leiden University). But little did I know how different it turned out to be.
Of course, the topics are different and such different programs also attract different kind of people, but I struggled with something else. It took me a while to name this struggle more precisely and it ultimately was about my own mental models and habits.
Based on my prior studies, I developed learning habits in which I almost always tried to identify the underlying structure and to reason my way up from there. I do not want to go into the debate about to what degree Computer Science is a formal science and to what extend it is an applied science, but undoubtedly involves a lot of math. Discrete mathematics, linear algebra, analysis, … the list goes on.
Focusing on the underlying systems and rules worked quite well for me with that. Although there certainly is always more to learn, learning feels somewhat like a ladder you can progressively climb rung by rung. Based on a foundation of mathematical theorems you reason your way up this ladder, but also the skillset you attain has more direct connection. Learning the basics of linear algebra is a necessary step to develop certain mechanism behind machine learning, analysis is needed to examine the runtime of algorithms and so on.
But now I was suddenly confronted with questions like “How does an effective government system look like?” which only lead to further questions like “Effective for whom?”, “What does effective mean?”, …
There just is no real foundation. Everything can be questioned. In fact, questioning almost everything seems to be the job of a good policy analysist for example. In my humble case (which is far away from being a real policy analysist or similar) I just was overwhelmed, still trying to climb a ladder which has no firm stand and ever-dissolving rungs.
Opposing to the image of the ladder, studying now feels more like investigating a gigantic cloud from all kinds of different perspectives. Topics like sustainability, comparative politics, local livelihood, and the philosophical question of free will are all interconnected, but hard to grasp. The cloud of such interconnections is ever moving, and it is hard to pinpoint individual topics because the context always matters.
Of course, knowledge in formal sciences is also very much inter-connected, but there is a precisely defined language to explore this connection. We call this language math. There are different fields of mathematics, but all have in common that they are based on clearly defined relations between conclusions (theorems) and assumptions (axioms).
Turning to the “real world” of social sciences I was unable to find such language. Of course, math is also widely used in social sciences in form of e.g. statistics or game theory, but it is always only somewhat descriptive. It is only a model and often far away from reality.
I had to accept that questions in social sciences cannot be reduced to precise underlying theorems. This did not surprise me, but it still took a shift in my mental models to adapt. Accepting that the underlying structures can only be determined to a certain point, means that you cannot arrive at certain conclusions.
This is where effective altruism sparked my curiosity, because the EA movement feels like the attempt to metaphorically put ladders into the cloud.
Effective altruism tries to reason clearly through topics which are highly diffuse. It does this with various methods always trying to depend on quantitative evidence and logical reasoning instead of speculation. Comparing e.g. the impact of distributing mosquito-nets against malaria and funding cataract surgeries is not easy and requires a lot of assumptions. But relying on detailed studies, cost analyses and impact comparisons helps indeed. Of course, this is not a new approach, but an especially interesting one regarding the questions asked within the EA movement.
It was a nice little “things fall in place”-moment when I thought about this metaphor and EA-like reasoning as the attempt of building ladders into the cloud.
A ladder-like logical structure is necessary to arrive at conclusions regarding the questions asked within the EA community. But we should keep in mind that all these conclusions – like “this NGO is the most effective” – base themselves upon ladders which were made up and which do not capture the real relations entirely.
Formal Sciences vs. Social Sciences: A Personal Reflection About Different Knowledge Structures
About 4 months ago I switched into “The Social Sciences” and got completely overwhelmed. I studied Computer Science at ETH Zurich before and anticipated that it would be quite different to study something like “Governance, Economics and Development” at a Liberal Arts and Sciences College (Leiden University). But little did I know how different it turned out to be.
Of course, the topics are different and such different programs also attract different kind of people, but I struggled with something else. It took me a while to name this struggle more precisely and it ultimately was about my own mental models and habits.
Based on my prior studies, I developed learning habits in which I almost always tried to identify the underlying structure and to reason my way up from there. I do not want to go into the debate about to what degree Computer Science is a formal science and to what extend it is an applied science, but undoubtedly involves a lot of math. Discrete mathematics, linear algebra, analysis, … the list goes on.
Focusing on the underlying systems and rules worked quite well for me with that. Although there certainly is always more to learn, learning feels somewhat like a ladder you can progressively climb rung by rung. Based on a foundation of mathematical theorems you reason your way up this ladder, but also the skillset you attain has more direct connection. Learning the basics of linear algebra is a necessary step to develop certain mechanism behind machine learning, analysis is needed to examine the runtime of algorithms and so on.
But now I was suddenly confronted with questions like “How does an effective government system look like?” which only lead to further questions like “Effective for whom?”, “What does effective mean?”, …
There just is no real foundation. Everything can be questioned. In fact, questioning almost everything seems to be the job of a good policy analysist for example. In my humble case (which is far away from being a real policy analysist or similar) I just was overwhelmed, still trying to climb a ladder which has no firm stand and ever-dissolving rungs.
Opposing to the image of the ladder, studying now feels more like investigating a gigantic cloud from all kinds of different perspectives. Topics like sustainability, comparative politics, local livelihood, and the philosophical question of free will are all interconnected, but hard to grasp. The cloud of such interconnections is ever moving, and it is hard to pinpoint individual topics because the context always matters.
Of course, knowledge in formal sciences is also very much inter-connected, but there is a precisely defined language to explore this connection. We call this language math. There are different fields of mathematics, but all have in common that they are based on clearly defined relations between conclusions (theorems) and assumptions (axioms).
Turning to the “real world” of social sciences I was unable to find such language. Of course, math is also widely used in social sciences in form of e.g. statistics or game theory, but it is always only somewhat descriptive. It is only a model and often far away from reality.
I had to accept that questions in social sciences cannot be reduced to precise underlying theorems. This did not surprise me, but it still took a shift in my mental models to adapt. Accepting that the underlying structures can only be determined to a certain point, means that you cannot arrive at certain conclusions.
This is where effective altruism sparked my curiosity, because the EA movement feels like the attempt to metaphorically put ladders into the cloud.
Effective altruism tries to reason clearly through topics which are highly diffuse. It does this with various methods always trying to depend on quantitative evidence and logical reasoning instead of speculation. Comparing e.g. the impact of distributing mosquito-nets against malaria and funding cataract surgeries is not easy and requires a lot of assumptions. But relying on detailed studies, cost analyses and impact comparisons helps indeed. Of course, this is not a new approach, but an especially interesting one regarding the questions asked within the EA movement.
It was a nice little “things fall in place”-moment when I thought about this metaphor and EA-like reasoning as the attempt of building ladders into the cloud.
A ladder-like logical structure is necessary to arrive at conclusions regarding the questions asked within the EA community. But we should keep in mind that all these conclusions – like “this NGO is the most effective” – base themselves upon ladders which were made up and which do not capture the real relations entirely.