The consensus is that there was steady, but slow growth. A few outliers doesn't change that.
You'll note that I specifically said "before agriculture," i.e. before the -10k B.C. point in your list. Following the development of agriculture there was of course substantial population growth over time, but before that not so much. I repeatedly noted that I was specifically claiming that
pre-agricultural populations were relatively constant (once all of the continents had been colonized) and merely that while there was population growth after the invention of agriculture it was slow and not necessarily very steady, with periods where growth stalled.
For the sake of argument, let's use the HYDE project, who estimates world population at
- 10.000 in -70bc
- 2m in -10k bc
- 18m in -5kbc
- 188m in year 1 ad
- 295m in year 1000
- 6145m in 2000
Just because you have exponential growth from 1000 to 2000 ad, doesn't mean that there wasn't substantial growth before that.
In fact I
don't accept using the HYDE project "for the sake of argument". The very fact that you have to pick a specific estimate, when there are many estimates out there with sometimes substantially varying values (for example, populations as high as 300 million for 1 A.D.) shows that this is not reliable evidence that the population has inevitably grown by a substantial amount every century and millennium since humanity evolved. It certainly does not prove that there are no periods when there was global regression in population, as you claimed.
Also, the very long timescales involved show that even taking those estimates at face value there was
not substantial growth over long periods. In fact, growth was quite slow! Like I show below, substantially below 1% per year, or even 0.1%.
Which, if you believe your own argument, would mean that you can't claim that there wasn't growth
That misunderstands statistics rather thoroughly. The null hypothesis is (normally, and certainly in this specific case) that nothing happened, and the job of the evidence is to prove that it
did happen. If the error bars overlap, particularly if they substantially overlap, then there is no way to reject the null hypothesis and therefore no reason to believe, on a statistical level, that anything happened. In other words, what they prove is that there is no evidence of growth occurring. They are
not evidence that there
was growth occurring.
In any case, I am not denying that there was growth on very long timescales, just that there were substantial periods where it is
not clear that there was any substantial growth between one tick and another--again, the 1 A.D. and 1000 A.D. examples are particularly instructive here, because the large error bars make it plausible that there was actually a
higher population in 1 A.D. than in 1000 A.D.
From 2m to 295m is hardly "sluggish".
It is if it takes
eleven thousand years for that growth to take place. That's a growth rate of just 0.045%. That's sluggish by any reasonable standard--over the course of a century, the global population would only increase 4.6%. To put it another way, if at the beginning of one century there were one hundred million people on Earth, at the end of it there would be about one hundred and five million. Barely noticeable, spread out over the whole planet. Putting it differently, if that growth rate had continued, we would be looking at reaching our current
actual population in about six thousand years. Slow!
Which is the point - why should we believe that we can predict what technologies will exist 1300 years from now?
Most unlikely, it will be stuff that we can't even imagine.
Well, we do have a rather better understanding of the nature of the universe than they did in 500 A.D.; there is a far greater chance that the things which we say can't exist actually cannot, physically, exist (just like the example I used of the perpetual motion machine of the second kind). But as I said the actual point is that saying "oh, there might be magical technology that would enable trillions of people to live in North America" is not particularly helpful for estimating the number that could actually live there
now, when such magical technology doesn't exist.
The population merely fluctuated regionally (and in the case of North America, locally) between 1 and 1000 AD. You can find huge distinctions, like for instance compare Germany in 1 AD to Germany in 1000 AD.
We are specifically discussing changes in global population, not regional or local populations. If the global population remains constant and is merely redistributed from one place to another, that still means there was no global population growth.