What's the smallest amount of text can I steal/spin that would be considered unique content? And does this bit of text need to be unique to the article or the internet in general? Follow me on this: I grab an article from Ezine*Articles.com and slap it onto a website. 100% NOT unique. I grab two articles about the same topic. (There's a TON of articles that are basically about the same thing on there.) I use the first half of article 1 and the last half of article 2 and make a new article. Would I now have something that is 50% unique to itself, or 100% NOT unique to the internet? Let's go nuts - I grab 1 sentence from 16 articles and make an article from that. (Instant Article Wizard does just this.) Google would've seen each sentence on the Internet, but not together in this order. This article is totally unique onto itself, but the parts aren't unique to the internet. What does Google think of such a creation? I go even further and change the middle word of each of the 16 sentences above. Each sentence is now unique, but the first and last word groups of themselves aren't. Is this now totally unique or does Google see that it's comprised of 4-5 word groups that aren't unique to the internet? At some point you can't break down the English language any further because certain short groups of words normally go together. How can you be penalized for that?