In How We Know, Chapter 6, p. 197, Harry Binswanger writes:
Knowledge is based on the data given in perception. From that base one builds new knowledge upon old, in an incremental, step-by-step process. Knowledge is not gained by revelation from on high and it is not gained in huge gulps. Knowledge is built up, in “crow-friendly” steps, from specific perceptual observations. We start somewhere: 1) where we are in the universe, and 2) with the kind of information our perceptual system is scaled to detecting. E.g., an infant lives in a specific location, with sense organs sensitive to a specific range of energy differentials and not others. He can perceive the people and furniture in the room, but not the atoms that compose them and not the ultrasonic frequencies that a dog can hear.
From his first perceptions, the child’s knowledge grows step by step. His progress is made possible by further observation, observation of new things and of new aspects of old things. At a certain point in his development, he is able to form concepts. From then on, observation and conceptualization reinforce each other, in a spiraling process, as I discuss later in this chapter.
Now we are prepared to revisit the issue of fallibility. When a child forms his first concepts, the ones for perceived objects, though some mental effort of directed attention is required, the process is infallible; there is, for instance, no such thing as getting the concept “table” wrong. Even if a child forms a concept of “table” that differs somewhat from the concept that adults name by that word, the child is nonetheless aware of whatever similarities he is aware of, and his condensation of the similar concretes cannot be faulted.
His discussion seems a bit muddled. A person may be wrong when he thinks that things are similar in some way. A child might think that a particular object could function like a table and hold a drink in a glass when it would fall over if you put the glass on it. I don’t think you can make a hard and fast distinction between how you group perceptions of objects and other ideas you have about the objects’ properties that might be wrong.
Elliot says:
The CR view on assumptions and foundations is that we can start anywhere. We can start with high level ideas or low level ideas. We can start in the middle. Anything goes because we aren’t trying to solve the problem of avoiding error by limiting where we begin our reasoning.
In Understanding Objectivism, Peikoff says:
Philosophy is a hierarchical structure. Each step is built on the preceding step. I’ve compared it to a skyscraper with fifty stories, and you have to know before you study any one floor or any one window how high up in the skyscraper you are.
He also says (bold added):
We are now going to do it in one specific field, and that is philosophy. Like all knowledge, philosophy has a hierarchical structure. To prove any particular item within philosophy, you must discover what it depends on. And to understand it, to chew it, you must know what context you are counting on.
The approach Peikoff describes seems like it represents an area of disagreement between CR and Oism, at least to the extent you take Peikoff to be treating the hierarchical structure of philosophy claim as an epistemological issue related to validating knowledge and not just as some useful way of organizing knowledge.
Yeah there’s disagreement there.
I think the answer is that there are many different hierarchies that work. So you can start anywhere and figure out something that works. Some starting places and paths are more convenient than others. But there certainly isn’t just one good one. So the truth is more similar to CR, but Oism is right that when you’re learning you need to have some kinda reasonable organization you’re using (or at least figuring out as you go).