For me, garbage is preferred. Trash is the ultimate expression of reality. It is the data spewed forth from our existence. From that data, you can back out facts about said existence.
I'm about a third of the way through Hillbilly Elegy, and yes it is a steamer, but that's exactly what makes it so enlightening.
It's a raw, unfiltered spew of verbal diarrhea chock full of the corn, emotions, contradictions, and mindset of the author. It hasn't been polished down into a coherent, rational, deductive set of reasoning. There's more entropy, more information, more substance than just a simple argument or streamlined story being told. It's a raw, rugged, and jagged slice of the author's mind that hasn't been commercialized so that "anyone" can understand it.
And memoirs like that are exactly what are needed to piece together a complicated tale. It's data, not just a model.
Data will arrive filtered through a range of systematic biases and random errors. It will be twisted by the instrument's (author's) personal mental structure. And that means that it tells you a lot about that structure.
Models are idealizations of the real world. They are not actual measurements or observations. They have been backtracked to try to figure out the reality that created those measurements. All models are wrong, but some are useful. They are generalizations used to describe what happens in the real world. They are a description of reality. Some can be very useful, because they can be very descriptive.
But data, perception, whether it's your perception or the perception of a scientific instrument, has been filtered into reality. From that data, you can build up a model. But the data is what is actually being perceived, and it is the record of what has actually happened.
When you learn from models, you build your understanding of how the world "really" is from a theoretical, rational, realistic viewpoint. When you learn from data, you build your understanding of the things that you see happening around you and how to deal with those events. If you only look at the data, you won't understand the hows and whys. If you only look at the models, you won't see obvious problems that are staring you in the face and you become disconnected from the reality of your perceptions. Indeed, your own perceptions become warped if you do not fully understand how to deal with data, and your understanding of reality becomes warped if you do not fully understand how to relate data to reality.
A major problem with our world today is that people are too dogmatic with their pursuit of either data or model, while rejecting the opposite side of the science of interpreting reality.
I love data. I was raised in a conservative, data-oriented world. I love taking data and trying to create some model of reality from that data. However, some people don't take the step of trying to connect data to some model of reality that their brain can comprehend and describe. Some people become too concerned with JUST the data at the expense of theory. They stop caring about what's really going on in the world. They only care about pursuing the data that makes them feel good and demonizing the data that does not. This, I feel, is the root of the cancer within the conservative movement. Conservatives feel, and forget to think about why they feel what they do.
My sister went the other way. An avid reader and active on social media, she fell into the trap of creating fantasies and conspiracies - the fun way of describing data and perception. Someone who is overly concerned with their theories will only accept the data that supports their theories, and will reject all others. They are more concerned with maintaining their world view and filtering their perceptions and the actions of themselves and others to impress their models of how the world should work onto reality. This, I think, is the root of the cancer within the liberal movement. Liberals think, and forget to check the world around them and if their theories are still connected to observable fact that can be sensed (felt).
But I digress. Hillbilly Elegy is garbage. It is a thrown together, unmottled collection of data about growing up as a conservative hillbilly. It is raw data from which you can construct and test any model you wish about the mindset of the author, and from there extrapolate to people like him. Although I have not read White Trash, from your analysis it sounds more like someone else's refined model. Without being able to compare that model to true data (in this case, data from the mind of an actual hillbilly, or interactions with hillbillies), you cannot test that model. I would rather build up a model from data and then use that model to predict future data and modify it as needed. I would rather not be served a neat, cleaned up model and be told "this is how the world works". All models are at least slightly wrong, and a more organically grown model that has arisen from the data will always fit the data and produce better predictions than one that has been constructed.
While getting my two BSes in Physics and Math, I learned that constructed models are good guesses when venturing into unknown territory, but you will always find surprises. This has been the history of all of physics, and science in general. Constructed theories are good bases to build from, but the data is reality and you must adjust your mental models to fit the data.
Human culture and psychology is so incredibly complex that data is even more valuable. Models that our brains can put into words are overly simple by comparison. For a model of something so complex to be remotely accurate it must be built up by organic means rather than constructed by simple rules. There are simply too many rules to keep track of, and each are more or less important depending on context. To truly understand these nuances, you MUST collect data. A lot of it. As the complexity of a model increases, the more data you need to constrain your understanding reality. This is a simple mathematical fact: to fit a more complicated model you need more data. A Hillbilly Elegy is nothing but data about poor whites, a group that is largely misunderstood and oversimplified by our society, and as such I feel that it is hugely important to the communication of their condition.
Either that, or I'm just a hillbilly conservative who loves data and that's why I prefer it to a more model-based approach. But I do believe that having your ideas be grounded in the truth (data) is the easiest way to stay grounded to reality. This explains the focus of most scientists on comparing theories to empirical reality and the focus of most historians of reconstructing history from primary sources rather than just tales, myths, and official reconstructions.
Or whatever, who really knows anything, yanno?