"As the dietary saturated fat content increased, liver pathology scores and ALT values decreased (P < 0.05). At 30% dietary saturated fat, ethanol-induced hepatic necrosis was eliminated, and micro- and macrovesicular steatosis and inflammation were markedly reduced, even though the carbohydrate and total fat content of the diets and the ethanol dose were identical to those administered to rats fed the low-carbohydrate, unsaturated fat diets."
This is pretty interesting too: http://www.ncbi.nlm.nih.gov/pubmed/21749725
Anandamide can be elevated from long-term consumption of high-omega 6, low omega 3 diets; it's a cannabinoid and and can produce - the sugar-craving munchies, just like the real thing!
(There's a cool discussion of the research by Emily Deans here: http://evolutionarypsychiatry.blogspot.co.nz/2012/09/omega-6-obesity-and-endocannabinoids.html
Anyhoo; what to recommend?
The standard Paleo recommendation from the most reliable sources (Paul Jaminet, Kurt Harris) is to limit PUFA to 4% of calories. If you eat about 2/3 of your energy as fat, that fat should be about 6% PUFA on average. If half your fat comes from dripping, dairy and coconut, the other half from red and white meat, olive oil, fish, nuts, seeds and greens, you'll come close enough to this magic figure; but it does kind of rule out taking 10mls of cod liver oil every day of your life.
I figure that we can count PUFAs in whole foods - fish and nuts - as being at least twice as beneficial and half as harmful as their purified triglyceride oils. No exact figures (except for krill oil), but good evidence that this type of difference is real. Peanuts? a) not a nut, b) source of the liver carcinogenic aflatoxins, and worth avoiding. Almonds and brazil nuts, black sesame and ground flaxseed are good sources of minerals, vitamins and prebiotics and worth having in the diet now and then.