Research into red wine's health benefits demonstrates the challenges in food and nutrition science. First, most of the findings have come from experiments on lower-level organisms, including yeast, worms and flies. The amounts of resveratrol involved would kill a person who drank the equivalent in red wine -- 60 liters a day, in one study done with mice.
Findings from research on human populations have been inconsistent or inconclusive. For example, one study comparing the health benefits of different types of alcohol couldn't discount the influence of its subjects' diet and exercise habits on the results.
Also, food and nutrient interactions are complex, and our understanding is evolving daily. Just look at how recommendations on fat intake have changed over the last 10 years, from a phobia of all fats to warnings about saturated fats to a focus on trans fats. The impact of a person's environment and genetics on the way his or her body processes food remains a mystery, but could be more important than previously thought. Regarding antioxidants, scientists aren't even sure they're asking the right questions and looking at the right relationships between causes and effects.