By now, you will probably recognize a pattern. To answer ontological questions, one must first consider reductive answers. If those fail, typically because there are counter-examples, one can still resort to primitivism. Giving a reductive answer to the special composition question means to propose a criterion for composition. Note that such reductive answers are moderate because they basically say that sometimes objects compose and sometimes they don't. Let's consider, for instance, contact as a criterion. Xs compose Y, if and only if they are in contact. At face value, contact is a reasonable candidate. The proper parts of the cello are in contact, so they compose the cello. We can multiply the examples. However, there are also counter-examples. Together with seven other planets, the Earth is part of our solar system, and yet these planets are not in contact with one another. This also applies to subatomic particles that together compose atoms. Contact doesn't seem to be a necessary condition of composition. It doesn't seem to be sufficient condition either. If the two of us, you and I shake hands, we are in contact with one another, but we do not compose a further object, do we? Similar objections can be raised to other reductive answers. Whatever criterion for composition one considers, there will always be examples and counter-examples. Ned Markosian also believes that composition sometimes occurs and sometimes it doesn't. In this respect, he sides with the moderate answers. But unlike the reductive answers that we've just considered, Markosian also believes that when composition occurs, this is just a brute fact, a fact that cannot be analyzed further. According to what he calls his brutal view of composition, there is no true, non-trivial, and finitely long answer to the special composition question. Let's consider his explanation of why the answer cannot be finitely long. Quote, "Suppose that there is no rhyme or reason as to when composition occurs and when it doesn't, as the view in question suggests. They could still be a truth of the form necessarily for any Xs, there is an object composed of the Xs, if and only if, it would just have to be an infinitely long list of every possible situation involving some Xs that compose a third object." End quote. Finally, let's consider van Inwagen's answer to his own special composition question. It's also a moderate answer because it puts forward a criterion, though a very special one, to tell cases in which objects compose from cases in which they don't. The basic idea of his mereological organism is necessarily for any Xs, the Xs compose, if and only if, their activity constitutes a life. You probably wonder what it is for the activity of some Xs to constitute a life. But according to van Inwagen, that's an empirical matter altogether. It has to be settled by biologists, not by philosophers. Note how peculiar van Inwagen ontology is, the only objects that exist in the world are mereological symbols and living creatures. Let's take electrons as mereological symbols, electrons exist. We are saying as well since the activity of our parts constitutes a life, we exist. However, in the world, there are no tables, planets, solar systems, and so on. These things only appear to be composite objects, but in actual fact, they are merely symbols being arranged table-wise, planet-wise, solar system-wise, and so on.