I'm not sure if you are actually disagreeing with TSG.

Dawkings repeats several times in the book that to him evolution is not about individuals, but about genes. Heck, even the title is "selfish gene", not "selfish tortoise".

Your tortoise benefits from screwing over his competitors, but his genes might not. Like, for example, suppose there is a combination of genes for noncooperative tortoise. He leaves others to die so the next generation is all assholes like him and they die out because there is noone to help when they inevitably get in trouble. Boom, genes are gone.

In fact, TSG also tries to analyze how cooperative and noncooperative strategies may coexist in single species and tend towards some equilibrium, which depends on costs/benefits of particular behaviors. And, iirc, such crude strategies like "always defect" often ended up outsmarted by subtler ones.

Dawkins seems to define "selfish gene" as "one caring about its long term survival and spread", whether individual tortoises like it or not. Because that's what ultimately counts - contemporary tortoises are shaped by genes which survived millions of years by whatever means happened to be successful, which needs not to be defection.

And, btw, you may want to actually read TSG, it's pretty cool :)

I second my sibling poster. You are clearly a very smart person who has no qualms about confidently misrepresenting a book you haven't read. It's not a charming combination - acquire some rigour in your thinking (for your own sake, if nothing else).

TSG hypothesis is that evolution happens at the level of genes (or memes, or whatever the replicant unit is). Not individuals, not groups, but genes. The individual level and group level evolutionary effects we see emerge from this, since a gene's prospects become intertwined with those of its neighbours. It is not well-described as a western, individualistic view. The 'selfish' is misleading if the negative value of the word is taken literally. Genes have no agency. The neo-darwinian view is simple but subtle, and is not well taught in schools. TSG is a great introduction to this way of thinking.

I believe Dawkins would argue this doesn't result in an evolutionarily stable strategy. Refer to chapter 10 of the Selfish Gene, I think it directly answers the concerns you have.

I think you'll find it does more to address your concerns here than you're expecting. It even includes a decent amount of game theoretic discussion with a view towards explaining cooperative and noncooperative behaviors as well as a more general notion of an evolutionarily stable strategy.

The still mainstream evolutionary biology it describes does not require any particular relationship between genotype and phenotype, just that one exists. Polygenic traits are just as well described by it as simple Mendelian traits. It also accommodates a range of factors like plasticity, epistasis, and gene by environment interactions that complicate the relationship between genotype and phenotype.

Additionally, your understanding of "selfish" as genes which produce defectors in cooperative scenarios is incorrect. "Selfish" is a metaphor to describe the fact that to be favoured by selection genes must contribute to a phenotype that increases their own propagation (to the degree that they are not constrained from doing so, and issues like linkage and pleiotropy aside).

Furthermore I don't see how your turtle flipping example negates the TSG idea. (but I'm very far from an expert on the topic)

Assuming you have not, if and when you read it you will find that there are no easy holes to poke. Dawkins doesn't just defend the gene, he derives it and then explains how it is the unit of persistence and why it is the persistence that matters.