 Let's get started by finding out why observation alone is not enough to answer our questions about the world. Remember that questions play a central role in our endeavor to find out things about the world. All insight at some point had something to do with a question. A moment of thinking, ha, that's weird, I wonder why this is so, or how does this work, or what if? From the previous chapter, you remember that some questions are harder to answer than others, and that there are questions where we are doubtful if we can answer them at all. Therefore, it makes sense if we try to formulate our questions in a way that gives us a good chance to answer them. Sometimes, we ask metaphysical questions. Those are questions about whether certain things exist in reality. For instance, scientists wonder if there is a soul and how it can manifest physically. So the question is, does the soul exist? Sometimes we ask normative questions. Should people who evade taxes be punished? Should you lie? Are there lies that are permissible because they provide greater good? These are questions about what is right and wrong, good or bad, and what should or should not be done. Some questions we ask are those that try to determine features of certain things. Which shape does DNA have? How tall is the European population? What features are included in the decision process when you decide whether to buy red or green apples? Those are descriptive questions that try to describe something very well. Mostly, the kinds of questions scientists are interested in are those of the type, does A cause B? For example, does taking vitamin supplements regularly make it less likely that people get a cold in the fall? In other words, does vitamin supplement ingestion cause reduced levels of the common cold? Why is the sky blue? Or in other words, which feature of the sky causes it to appear blue to us? Those are questions that deal with causal relationships. They deal with making precise predictions of what happens to a thing if something that is causally related to it changes. They also deal with explaining how these mechanisms work. What kind of questions you ask has to do with the philosophy of science that you follow in which area of science you're interested in. What do you believe the relevant questions are? And what do you believe are questions that you can answer? What would a good answer look like for the question that you are asking? The types of questions you ask and what you think is a valid answer to them will often also determine which methods you use to answer them. Normative questions are those that philosophers and legal scientists might ask. They usually deal with interpreting what we think makes sense, with philosophical and moral rules and how they should be applied to different situations. Often these sciences rely on verbal arguments and on thought experiments and on logic. Natural scientists are often interested in causal types of questions. They often want to predict what happens when we do something and come up with technologies, strategies and mechanisms to control the world that we live in. Thought experiments and the natural sciences are not quite enough. For this type of question, you need observations of real life. The idea is to see if the predictions you make work out and if the explanations you have of relationships can indeed be found in the world. In the context of medicine, for instance, causal questions try to distill the causes of illnesses to basic biological, chemical and physical functions. Natural laws that predict why and how and when someone will have a certain illness. In psychology, we try to predict behavior, feelings and thoughts as driven by personal and situational factors. We try to make these causal claims based on empiricism. We observe if the things we predict happen. Where do these questions come from originally? Usually, also from the real world. After all, if you haven't seen anything, how can you ask yourself how it works? Or why it works the way it does? By seeing things, what I mean is that you observe the world around you, that you listen to and read about what others think about the world. Or did you learn about scientific evidence about something and think further about it? Then you formulate speculations about causal relations. Why do you think X happens? You speculate, maybe because of Y. Then you go into the real world to observe if this is really the case, or if it isn't. Let's try an exercise about asking questions and coming up with speculations that might be interesting to investigate in the real world. Let's begin with this exercise. Pick an object in your immediate vicinity. For me, that's my woolly hat. What's your object? Describe it and its functionalities or activities. What are you looking at? What does it look like? And what does it do? Next, ask some more questions about the object, its looks and its functionality. Why does it look the way it does? And how does it do the things it does? Finally, try to come up with some speculations or hypotheses to try to answer the questions you asked yourselves in the previous step. Take some time to complete this exercise and then return to this video so we can move forward together. Okay, now let me tell you more about my hat. My hat is black and it's made from wool. It's been knitted from wool in a shape to fit an average adult head. It's a one size fits all item and it has an artificial fur bobble on top because that's fashionable right now and buying real fur is considered cruel. It's winter here at the moment and it's quite cold outside so I have this hat with me to keep me warm. The function of the hat therefore is to help me preserve a comfortable body temperature despite adverse outside conditions. Next, some more questions I could ask about this hat. Why does it keep me warm? Why is wool a good fabric for preserving warmth? Why is it not fit more tightly? Wouldn't that help keep me warmer? I think that's an interesting hypothesis. I wonder if my hat would keep me warmer if it fit more tightly. Next step, speculating. I think that the fit of the hat determines how warm it'll keep the wearer. In other words, my prediction is that tighter hats make people feel warmer. All right, so how do I find out if this is true? Let's not try to think of an observation we'd have to make to address our speculation. Which circumstances would I need to observe to see if my speculation is wrong? In the example of my hat, let's focus particularly on the last question I formulated. Wouldn't a more tightly fitting hat keep me warmer? What I could do now is observe. I could go outside on a cold day and ask everyone who's wearing a hat how warm their head feels. I'd distinguish how tightly fitting the hats are by looking at them. Then I would know if people who wear a more snugly fitting hat feel warmer than people who wear a hat that fits only loosely. Here's another exercise. Let's see if we can find ways to answer our questions about your object. How would you answer your hypothesis about the object by observation? Pause the video and think about it. Now, that seems like a good way to go about answering our question, right? So why would you even think that this kind of observation alone could not be enough? Let's go back to the example of my hat. I wanted to go out and ask people how warm they feel and compare if people who wear more snugly fitting hats feel warmer. What could go wrong? Quite a number of things as it turns out. People's hats may differ in more ways than just in the fit. For instance, some people may have hats made of different materials or knit more loosely so that the specifics of their hat may impact their hats' ability to keep them warm. Also, it may be windier or colder when I ask one person than when I ask the next person. Finally, the people themselves may have different characteristics. Maybe people who have longer hair feel wear-protected against the cold by their hat already and think that their hat makes no big difference, although someone who's bald would disagree. Do people also wear a scarf? Maybe a scarf-hat combination is extra effective in making people feel warm. So the external circumstances in which I conduct my survey may influence how effective people feel their hat is keeping them warm. I'm sure you could come up with more examples of things that we could miss. Now, revisit your own example from earlier. Pause the video and think about which problems you might encounter while trying to find an answer to your question using observation only. So what we have done is shown a number of alternative explanations for why people may feel their hat is more or less effective in keeping them warm. Other than the one explanation we were originally interested in, that is the fit of the hat. What we want, though, is to exclude alternative explanations so that we can isolate a certain explanation of an effect. We want to make sure that we find out what the influence of one thing is on another. We want to make sure that what we learn informs us about exactly that relationship and is not clouded and obscured by other influencing factors. Isolating a certain relationship is tricky when you only have observation as a tool, though. Why? Well, because there are hundreds of possibilities for alternative explanations. We would have to record everything about the situation and try to find one that is exactly comparable to another. And just from our little exercise before, you can tell that this will be immensely difficult, if not sometimes impossible. That's why we often rely on experiments where we have these things under control. Nevertheless, there are things that you simply cannot do in an experiment. For instance, you can't assign people to different gender or to different nationalities or personality traits at the snap of your finger. To study such categories, we therefore have to rely on naturally occurring variation and we cannot quite run an experiment. So there are some limits to where you can do experiments and sometimes observation is all we have. In such cases, we have to be extra careful about the inferences we make. In this part of the course, we have exercised our ability to ask questions about our surroundings and to make speculations about the world. We have also practiced coming up with ideas how we could answer our questions using observation. At the same time, though, we have seen that observation alone is not always enough because it cannot guard us against alternative explanations. Alternative explanations make it hard or even impossible to say if the question we were originally interested in has been answered properly. And by properly, we mean that we isolated the relationship we are interested in and make sure this is indeed the question we are addressing without clouding our answer with noise and error. Sometimes, though, observation is all we have. In the next part of the course, we will take a look at what we can do instead of only relying on observations, running experiments.