Even without read-the-thermometer records, tree corings and lake bed sediment samples and a number of other go-out-and-look methods can get pretty useful atmospheric data from the past. I don't know what the resolution would be, but it might fill in some gaps.
One challenge with using things like trees to estimate long-ago temperatures is that you have to build a complex mathematical model to map to temperature, and I'm aware of at least one famous study that got the math wrong, to the point of being potentially suspicious. (But I haven't studied this stuff enough to know whether or not other proper studies exist)
> One challenge with using things like trees to estimate long-ago temperatures is that you have to build a complex mathematical model to map to temperature
As it turns out, that's what you have to do with fairly recent temperatures measured with thermometers that aren't located in the same places, too.
It has been pointed out that tree rings respond mostly to changes in precipitation rather than temperature. If you can properly compensate for that, and for various other potentially confounding factors, then maybe you can derive some proxy temperature data from them. Other proxies have their own problems too, of course.
That's the project that was started by a physicist that didn't trust the numbers previously in the field, and basically called climate scientists amateurs, and set out to show them how they're all doing it wrong.
What happened in the end was he just recapitulated exactly what climate scientists had been saying all along, so he had to trumpet "oh but it's never been done this precisely," just glancing over all the damage he had done to the credibility of the field with his completely unfounded doubt.
Replication is great, and I'm glad Berkeley Earth is still at it.
But we need to start holding people accountable for their predictions on this matter. If somebody is filled with unfounded doubt, why the hell should we listen to them until they've actually done the work to provide data?
If you haven't even bothered to check sources but want to parade around how you think the data is untrustworthy, data which you haven't even sourced much less evaluated, why should we listen to such data-free and evidence free speculation that has proven to be wrong every time is has been evaluated?
If somebody is skeptical of climate scientists at this point in history, it says a lot about their inability to evaluate technical matters and their ability to form unfounded opinion when there is no need for unfounded opinions. That reflects extremely negatively on the technical capabilities of a person.
That's mighty big talk for someone who probably has done little to no due diligence on the data themselves. But I have, and here's just some of what I've found, specifically in reference to Berkeley Earth (BEST):
1. The initial BEST report (I haven't really looked at later ones) had some remarkable things buried down in the footnotes which, if taken at face value, put everything else they had to say in serious doubt.
2. If I go to their web site and look at the temperature data for my local area, it's obvious that the first 100 or so years of that data is basically fake. I happen to know from previous research that there are no local temperature records going back that far, and if you pay attention to the details they admit as much themselves.
3. Furthermore, I've lived in this area for over 40 years now, and during that time it just so happens that I've experienced both near-record highs and lows for the area - the hottest and coldest that I've ever been in my life. Yet those highs and lows don't show up in the Berkeley charts, which leads me to believe that the whole data set that they're using is basically garbage.
4. If I generate a "Stripes" report for my area, it looks little to nothing like the regional/global stripes reports currently being bandied about.
5. Last but not least, not long ago a researcher publicly pointed out (in considerable detail) the generally remarkable lack of quality control of various climate data sets, on which Berkeley (among others) depends. And he was immediately attacked by Steven Mosher, one of the main Berkeley guys. But Mosher was so anxious to try and discredit the guy that he apparently didn't read the new report too well before trying to point out "obvious mistakes" in it. For example, he (Mosher) used the wrong date ranges, the wrong filtering criteria, and so on in his critique - not the same ones that the report itself was actually using - and this was quickly pointed out to him. (The "problems" that he pointed out mostly went away if you used the correct criteria.) So Mosher was the one who ended up with egg on his face, and I don't think he ever properly apologized for this, either.
But rather than just take my word for such things, you should probably go do some looking for yourself.
Go here to look at Berkeley temperature charts for your area; read through all of them, and pay attention to the details:
http://berkeleyearth.lbl.gov/city-list/
I'm not saying it's fake, but I'd like to know where the data comes from, and how reliable it is.