I agree with you about it being a useless tag because it is as you say supposed to point away from duplicate content to a master page, they should remove it.

How long have they been using it?
Is the Content still indexed ?

Google would probably just ignore it, but who knows. What if it ended up in some sort of loop where it keeps reading from the top of the page until it hits the tag again.

That's correct.

Yes it can and it's a common mistake. So you should get rid of those.

If you content is original there is no prob in adding it (make sure that rel="canonical" is the same url itself).
If your content is copied one you can give <link rel="canonical" href="http://Actual content location Itself.html> to the header of duplicate page.

The Aim of tag is to tell Google Bot the location of original content .

I have this exact same question and would like to know more as the information I've found thus far is conflicting.

Perhaps Yukon can chime in

Anyway, I was thinking of doing the same thing... adding rel="canonical" href="http://URL Of Itself.html as sort of "preemptive" strategy should my site be scraped (and of course, to avoid dup content issues).

I've found the following info on these industry "authority" sites:

Understanding Negative SEO & How to Defend Your Website - Search Engine Watch (#SEW)

-- they suggest "The use of an absolute rel=canonical tag on pages can help establish your site as the authority when content is getting scraped by a third party."

Rel=Confused? Answers to Your Rel=Canonical Questions - Moz

-- They say this practice is actually ok (see #5)

seo - Canonical link rel tag pointing to the same page - Webmasters Stack Exchange

-- They to suggest this is ok (ie: "There's no harm in doing so").

Not trying to contradict anybody... just would like a solid answer and perhaps one even based on quantified testing.

Thanks!

Thank aliceria, I think that much is clear. But the OP's question (and mine too) is whether or not it is harmful to reference origin page with rel canonical.

The canonical tag is supposed to point to the page/URL you want to show up in the SERPs, that's it's purpose, otherwise Google will make the decision for you (no canonical tag/s) which is a bad idea IMO.

Having multiple internal pages with the same content can't really hurt rankings, the problem is If you have duplicate pages & don't tell Google the preferred URL & Google decides to display a duplicate page while your busy trying to rank the original internal page that's buried in Supplemental SERPs (duplicate page overflow).

If you have duplicate pages either remove them from the site, noindex, or use a canonical tag the way it's supposed to be used.