From: To: Danny Hillis • Cc: "jeffrey E." <[email protected]> Subject: Re: Date: Sat, 03 Dec 2016 16:05:44 +0000 > Hi Danny, I appreciated your thoughts yesterday re "stories." Three follow up thoughts: I) In cases where we can't tell complete stories, like your sorting algorithm, we can presumably tell stories about certain features. And I guess this is what I was trying to claim about evolutionary theory; we can't hope to predict a priori which features will show up in which taxa or why humans evolved, but we can explain why sex ratios are about 50:50. And in such cases we can still distinguish between good and bad stories (the god explanation or a group selection explanation vs fisher). But I guess this raises two questions: 2) In such domains, which kinds of stories will we be able to tell and which kinds won't we? Should we be striving for a Darwin-like story of human culture? For a high level pseudocode-like story for how the mind works? Or are there other questions we should be asking? What is the level of analysis we should be doing in these domains of research? 3) And in such domains, how can we distinguish good from bad stories? This seems like a particular problem in domains, like the social and biological science, where even the good stories are incomplete and never 100% accurate. But there are still better and worse stories. Currently, as you pointed out, these fields seem rife with bad stories. How could we more systematically distinguish the good from the bad, in such domains? Is there a conceptual tool or a litmus test they might be able to apply? Anyhow, I look forward to chatting again. I hope you didn't find my comments or questions off-putting. I didn't get to hear what your daughter does, but it sounded like she has an interest in psychology. And in evolution. I teach and do research in both those departments, and would be happy to chat with her more, if she had an interest and found herself by Harvard again. I app