On my re-reading of The Analysis of Mind (1921) I was amazed by how much more I understood of Russell's lectures than on my first reading. Between my two reads I have read much of Russell's earlier work so perhaps this helped to appreciate his views on psychology much more.
Anyway, Russell's central thesis is pretty straightforward. He analyzes all the subjects of psychology, as the study of our mental life, and comes to the conclusion that all of this field can be reduced to sensations and images. Images themselves can be reduced to mnemetic causation (i.e. prior experience is stored and influences new experiences - images in this sense are stored copies of prototypical sensations). Mental occurrences like emotion, will, desire, belief and remembering are all complexes constructed out of sensations and/or images. Things like consciousness and mind are, for Russell, nothing but inferred entities and should be avoided as much as possible - at least when approaching psychology as a science.
In this regard psychology approaches reality nearer than physics, which deals with logical fictions such as matter and atoms, that are constructed out of systems of occurrences. A future, ideal physics should be construed out of particular occurrences and the causal laws guiding their dynamical development over time - as is already done in psychology.
Throughout the series of lectures one can see the contemporary scientific insights beaming through - behaviorism, psychoanalysis, pragmatism, William James' empiricism and realism, etc. Russell uses what he can but in the end builds his own theories of our mental life and all its aspects.
One of the more interesting parts of the book are the chapters on belief and truth and falsehood. This is because Russell has to offer a psychological explanation for his theory of knowledge. According to Russell, truth is a value of a propositional function, which itself is a complex deriving its meaning from the meaning of its elements and their relations. That is, all our beliefs are complex of a proposition and an accompanying feeling. The proposition contains an object and its relation to that object is its meaning. In other words, a proposition refers to some fact or system of facts outside the mind and is true or false in virtue of the existence or absence of this fact or system of facts. Ultimately, our beliefs (and thus our knowledge) consist of feelings and of words and/or images. And feelings, words and images are derived from sensations. In effect, Russell founds his theory of knowledge in sensations of particular occurences, as the ultimate constituents of reality.
Another very interesting part of the book is where Russell deals with the connection between psychology and physics. In physics similar particular events are grouped according to all the places where they appear; in psychology different particular events are grouped according to the place where they appear. That is, a physical object (say, a star) is the collection of all its appearances; and a psychological object (say, a sensation of that star) is the collection of all sensations simultaneously making up the perspective. Perspectives over time are called biographies and are what distinguish living from non-living matter. That is, organisms have a continuous series of perspectives.
This also illuminates Russell's critique of physics: a star is the system of all its appearances (whether they are observed or not), which means we are already dealing with a fiction, since the system itself is a construction out of the raw data that are the appearances. Ultimately, all there is is appearances (i.e. data) and psychology deals with data one datum at a time. In this sense, according to Russell, it approaches reality much nearer than physics. Causal laws are, for Russell, that which binds occurrences together through space and time and makes these occurrences appear as they appear. And this allows him to claim that in the future physical laws and psychological laws should approach one another ever closer. He is skeptical about whether a perfect unification will ever be possible, yet the ideal is good enough.
This book will be on my mind for a long time to come. So many gems of insights, so many unique perspectives, so much clear thinking. And notwithstanding the date of publication (1921) and the complexity of the subject matter it is perfectly readable still. Definitely recommended for anyone interested or occupied in psychology!