I was taught the concept using the water example in undergradudate, and find the analogy is spot on. Other comments are considering more complicated marginal effects or the intersection/equilibrium of multiple marginal effects and confusing the point.
I didn't claim anything such as this and I'm having trouble understanding what you are responding to in my post.
A utility function is just a value of utility received at different quantities and costs. They are just basic models for helping to simplify our understanding the world. There's no need to begin comparing expertise with utility functions. The author is talking about the utility he receives at the margin of one additional widget in his decision making process, and what that this type of scenario (when a commonly cheap item can, in our decision making process, can have a high marginal utility--ie quantity is only 1 and 1 is necessary for a finished product) means to us in our human experience.
"Right now this screw is worth exactly the selling price of the whole motorcycle ..."
This is not marginal utility as somehow the value of this screw is equal to the entire amount of the motorcycle. There is no marginal utility to be had because decreasing it or increasing it has no effect. This would not be a continuous function. You could talk about the marginal utility of one having a motorcycle or no motorcycle. It's not related to the increase in screws.
Additionally, to say that expertise in a subject is not needed is perplexing to me. So much so I literally can't begin to understand how you'd surmise such a claim honestly.
How much economics did you take? I ask because it sounds like your understanding is that from a class or two, and not the bulk of your education. I think if it was we would not be having this discussion or you'd somehow prove it instead of reiterating the same points.
I don't see it as productive to respond on the internet to the posturing and credentialing about who has seen more utility functions or taken more econ classes, as you wrote above.
I can assure you that Robert Pirsig garners utility at different costs and quantities in the scenario he described. That is all that a utility function is: a function of utility at varying quantities. This is an elementary topic and there's nothing more to be argued here. I'm sorry that explaining it again rubbed you the wrong way and encouraged you to write a condescending response towards me.