I have inherited a Sitecore architecture that is a mess from an SEO duplicate content POV.

There are multiple aliases that have been created (and indexed by the search engines) for many of the 2nd tier pages of the site.

Due to server issues, I am not able to 301 redirect these duped pages, so I would like to use the rel=canonical tag in an attempt to try and get Google/Bing to recognize the correct pages I would like to appear in the index.

I have blocked the most extraneous duped pages with a robots.txt file, however, since Google/Bing have already spidered many of the duped pages, I need to keep them accessible to the spiders, BUT removed from the index.

The catch is, since the duped pages are aliases (and don't really physically exist in Sitecore that I can find), I am not sure how to go about using rel=canonical - or if I even can in this situation..?