Let’s deconstruct this often made claim that systems thinking (ST) stands in contrast to (in this case on opposite ends of a continuum from) reductionism. This claim appears reasonable at first, and it is one made my many of the founding fathers of systems thinking as well as contemporary systems thinkers. But there are some dangers, mostly having to do with misleading people new to the field toward misunderstanding what it means to be a systems thinker.
First, if reductionism (breaking things down into their constituent parts) is on one side of a continuum, then holism (seeing things in the context of their whole) is on the other. If reductionism is the opposite of ST, then ST and holism are synonymous. Here lies the problem…
No matter how holistic one gets, there is always a larger whole to act as context, and one is therefore always surgically removing the whole under consideration from the whole it is a part of, which is one of the chief criticisms with reductionism. So being a pure holist who does not reduce or being a pure reductionist who is not holistic is absurd and it is a false dichotomy. We need to do both just to begin thinking about anything.
Therefore, to break things down into parts (reductionism) is an act of systems thinking as much as seeing things in the context of their whole. A healthy part-whole balance is a necessary aspect of systems thinking. There’s an old saying that there are two kinds of scientists: splitters and lumpers. Systems thinking then is the act of not accepting this duality and of being a "splumper."
Of course, this creates another problem. When we “break” things down into their structural parts, we are “breaking” the dynamical relationships, so we must therefore take extra caution to identify the broken relationships and think of them as the dynamical parts of the whole (or we may end up with unexplainable phenomena that we must label emergence).
Of course, now we have another problem: that even our most holistic thinking cannot be infinite and complete. Even holistic thinking must draw a boundary somewhere. At the very least between what we consider and what we do not. At that moment a distinction forms between what we decided to consider and what we decided not to consider. This boundary is a distinction where we identify some thing(s) as being inside the scope of our observation and some other things as being outside the scope.
This in turn causes yet another problem which is that the choice of a boundary is a subjective one, requiring perspective, so we must consider that these distinctions we make for the whole (and each subsequent distinction we make of the parts) is perspectival. Change our perspective and the distinctions we make change, too.
For a much more extensive introduction to these ideas see the new book, Systems Thinking Made Simple: New Hope for Solving Wicked Problems.