The arguments used in Brian J. Baker's article to "debunk" global warming are many, and some of them are in contradiction with each other. Apparently, there are lots of blogs and websites (most of them run by conservatives and even overt reactionaries) devoted to climate-change denial. Most of their authors use the same set of arguments over and over again, and the same stuff has been pasted in Baker's article. I shall try to reduce each argument to its basic core.
Argument 1: Maybe there was a lot of CO2 a long, long time ago (so what?)
Some data show that possibly the correlation between CO2 and warming was different hundreds of millions of years ago. No big surprise therefore that our models do not work for the Phanerozoic period, if one considers that the Earth was quite different at that time. For example, land plants did not yet exist. It is quite clear that land vegetation is rather important as a weather factor.
However, this point is an important one and is not a complete hoax, but it hardly falsifies present-day thinking on global warming. Working out climate and gas concentrations for periods so remote in time is very complicated, and different methods are used for different time intervals.  The result is the union of disjointed curves and anyway there is no lack of discrepancies between different data sets as this picture  shows:
Baker, of course, picks the most dramatic data and forgets to draw the huge uncertainty range (look at the yellow shading) in the early-Cambrian CO2 spike. His unsourced temperature data seem even more suspicious. These graphs come from a notorious pseudo-scientist and producer of hoaxes, E-G. Beck, but we will talk about him later. 
In general, Baker makes a curious usage of graphs in his article. For example, he adds a corollary: life flourished in these 600 million years and yet CO2 has been very high many times. Then, he shows a graph that proves exactly the opposite! Let us just take the graphs he himself uses and superimpose them:
Not an elegant diagram, but it illustrates a simple fact. The relevant curves are the red one (carbon-dioxide concentration) and the green one (thousands of Genera, a rough measure of biodiversity). Apparently, when the red curve goes up, the green curve is low and vice versa: if this data is correct, life flourishes when CO2 decreases as in the Cenozoic!
The correlation seems to be the opposite of what Baker was ineptly trying to prove. If that were true, we should seriously worry about the carbon emissions themselves and not just their effect on global warming.
Argument 2: In the Middle Ages it was quite warm too (so what?)
Baker compares two graphs issued by the IPCC, one allegedly from 1995 and the other one from 1998. In the first one, the Medieval Warm Period (MWP) seems warmer than now. In the second one, there is the "hockey stick", i.e. a long period with variations that show no large long-term trend followed by a clear upward trend in the last decades of the 20th century.
How can such a big change in the data be explained? Baker's answer is simple: they had to get rid of dangerous evidence. The truth is even simpler: the first graph (that comes from the first IPCC report of 1990) does not represent average global temperature but just the temperature of Central England. The excellent award-winning website RealClimate explains:
"It originally is taken from the first IPCC report of 1990 [...] At that time, no large-scale temperature reconstructions were available yet. To give an indication of past climate variability, the report showed Lamb's Central England estimate. (Unfortunately this was not stated in the report - an oversight which shows that IPCC review procedures in the early days were not what they are now.[...])" 
The story of this graph is quite complicated and has been explained in detail by comrade Emil Reed in a previous article. (Emil Reed replies to Brian J. Baker on climate change).
In fact, apart from the pollution that produces global warming, there is also a lot of rubbish denying global warming on the basis of local data produced every day, and it usually originates from the same greedy private companies! Also Baker's article is filled with at least eight charts based on local data (China, Central England, Arctic Pole...). In part three, we are also informed that "Kazakhstan was subjected to its coldest winter weather in 40 years". Inside my fridge it is very cold too and I did not notice any warming trend.
Global warming does not imply that every part of the planet will warm up. On the contrary, some opposite effects are expected on a local scale. This is the meaning of the word "average", as in "average global temperature". Global consequences of the climate change will nevertheless be felt almost everywhere, for example in the form of more frequent extreme conditions and rises in sea level.
Argument 3: Some say that the "hockey stick" is an artefact (but they are wrong)
«In the end, it's important not to lose sight of the forest for the trees, where the ‘forest' refers to the totality of scientific evidence for global warming.»
(Richard L. Smith, reporting on the hockey stick controversy in the newsletter of the American Statistical Association, Section on Statistics & the Environment, Spring 2007 issue )
This is another old story, but it has a much more interesting political side. Baker tells the story but, even though his article is mistitled "A Socialist Perspective", he fails to mention that he is parroting the arguments used by the Republicans in the US Congress in a grotesque "trial" against climate change occurring in June 2005!
The famous hockey-stick-shaped graph shows how global average temperature spiked upwards in the last few decades. A climatologist called Michael Mann plotted the first hockey stick graphs in 1998 and 1999. For this crime against the free market, he has been put under investigation by a subcommittee of the US Congress chaired by the Texas Republican Joe Barton, a global-warming denier and a pollution deregulation fanatic, with clear links to the oil industry. The investigation immediately turned into an all-out attack on Mann's work and professional integrity. The "network of ties" between Mann and 43 other palaeoclimatologists has been put under examination, and the whole palaeoclimate community has been accused of having a very bad scientific record. 
The approach of Barton and his stooges was clearly aimed at launching a witch-hunt on a part of the scientific community that was perceived as dangerous for the vested interests of his favourite lobby. If anyone is corrupting science in pursuit of their own political agenda, it is precisely these kinds of people!
The hockey stick graph has been replicated by several later studies trying to reconstruct the past behaviour of world climate. Some of their results are shown here :
The hockey stick part is the last millennium, since temperature began to decrease after the Medieval Warm Period until the latest global-warming phenomenon (incidentally, Baker writes that the MWP is "eliminated" in these graphs - he was probably looking at them upside-down like the biodiversity chart...).
The first important hockey stick graph  was generated by Mann and others on the basis of the features of tree rings (the rings you can see if you cut a tree's trunk). Some trees are pretty old, therefore it is a good way of working out the temperature of a place centuries ago. Of course, several sets of data had to be used to give a picture on a global scale; Mann and the other authors of the first hockey stick graph actually focused only on the Northern hemisphere.
What is the relation between tree rings and temperature? They tried to determine it on the basis of instrumental data - the temperatures we actually know for sure. Using well-established statistical techniques to extract a pattern out of raw data, they tried to "sum up" a big amount of data reducing it to its "principal components". This makes sense also as a way of filtering out parts of the data ("noise") that are not related to climate.
Of course, the reducing of the whole data set to a simpler description had to pass through a validation check: if the principal components, calibrated on the basis of 20th-century data, were still significant enough to describe effectively the 19th-century data, probably no fundamental information was lost in the process and they could be used to guess the approximate temperature of previous centuries on the base of tree rings. Mann et al found two principal components, which were good enough to mimic the whole data set and they used them to produce a hockey stick graph out of the tree data.
Two researchers, Stephen McIntyre and the market fundamentalist economist Ross McKitrick, wrote an article against Mann et al in 2005.  Climate change deniers see plots everywhere, with scientists being bought by powerful green lobbies (or capitalist lobbies, in the Baker variant of this conspiracy theory), but strangely enough they do not notice how Mr McIntyre signed his most famous article: "Stephen McIntyre - Northwest Exploration Co., Ltd., Toronto, Ontario, Canada". This man is not on the payroll of the oil and gas capitalists: he is an oil and gas capitalist! He founded Northwest Exploration Co., Ltd., to be taken over by CGX Energy Inc., who hired McIntyre as a strategic advisor. He also worked for the Canadian government and, more notably, is listed as an "expert" in the portfolio of the Marshall Institute , a lobby based in Washington, D.C. that campaigns for the Strategic Defense Initiative and space weapons, against climate change theories and in favour of the tobacco industry: warmongering professional deniers, hardly a useful source for "a socialist perspective" on imperialist wars, climate change or lung cancer.
Nevertheless, McIntyre, during a round-table chat organised by the very Marshall Institute, assures us that he acted only for the sake of truth:
"This is costing me money to do. Normally I would be working on some business deals. I spent quite a bit of time on this and I found it quite interesting. Fortunately I have had some stocks go up" 
Fortunately indeed! We all hope that global-warming hysteria does not make them go down!
Without going into details here, the article by McIntyre and McKitrick questioned a detail of the statistical technique used by Mann and some other flaws in the usage of some series of data related to North American trees.
McIntyre proposes another way of picking the principal components. He applies an alternative way, and then he takes two (different) principal components again and argues that the result is different. Mann replies that by changing the way of picking the principal components, the number of significant principal components must change too. The correct number with the alternative technique is 5, according to Mann. Using 5 components, the result is the same again.
This is all very interesting, but could take us a bit off track. In the end, what is politically important is to know if this controversy makes some difference. It does not. Flaws in the data can be fixed and the hockey stick would still be there. The whole selection of principal components could be abandoned and the hockey stick would still be there. McIntyre's method could be utilised and if relevant data are not excluded the hockey stick would still be there. Methods that are different altogether can replace principal components analysis and the hockey stick would still be there.
Also graphs published by sceptical lobbyists like the Fraser Institute  cannot help displaying a hockey stick:
Argument 4: Using different methods to plot the historical graphs, there is nothing special happening recently (just not true)
Comrade Louis Proyect, who maintains a successful mailing list about Marxism and an interesting website, had it right when he pointed out the light-minded usage of quotations by Brian J. Baker.  Quite often he seems to have cited rumours and allegations more than actual statements. In three different cases, he assures us that there are alternative views that deny, or denied in the past, the global warming postulated by those blind believers that just use the silly graphs produced by mainstream dendro-chronology (study of the past through the examination of tree rings). But also these citations are not very scrupulous.
Craig Loehle should be one of the alternatives. He proposes a different opinion on the way tree rings depend on climate. Brian believes Loehle's view to obliterate his own position, but Loehle seems to be uninformed of that, because as late as March 2007 he co-authored an interesting article called Forecasting the Effects of Global Warming on Biodiversity .
Another candidate for leader of the denial movement, in Baker's opinion, is Shaopeng Huang. As Proyect suggested, he is hardly in disagreement with the scientific consensus on global warming. He is the custodian of the Global Database of Borehole Temperatures and Climate Reconstructions, available online . The idea supporting his research is that "Departures from the expected increase in temperature with depth (the geothermal gradient) can be interpreted in terms of changes in temperature at the surface in the past, which have slowly diffused downward, warming or cooling layers meters below the surface" . So, according to a group of researchers, borehole data might be regarded as a buried memory of long-term trends of temperature. In the abstract of one of his 1999 articles  Shaopeng Huang's findings are summed up in this way: "The results confirm the unusual warming of the twentieth century revealed by the instrumental record, but suggest that the cumulative change over the past five centuries amounts to about 1 K, exceeding recent estimates from conventional climate proxy". In any case, Shaopeng Huang is living proof that the hockey stick graph is confirmed by independent studies using a completely different way of deriving the temperature trends.
This is the global conclusion derived by 952 individual reconstructions based on borehole data, as shown on the Global Database of Borehole Temperatures and Climate Reconstructions:
The third mythical denier, according to Brian Baker, is Dr Stephen Schneider - not now, but 30 years ago. Now Schneider is one of the most famous supporters of the theory of global warming, but Baker says that "Before Global Warming became the politically correct scientific fashion of the 1990s, [...] it had become a scientific article of faith that the Ice Age was about to happen [...] Schneider wrote paper after paper describing the inevitability of the coming ice age".
This is a widely circulated double myth. First of all, Schneider did not write "paper after paper" predicting an "inevitable" ice age. He co-authored one paper, whose main author was S. Ichtiaque Rasool, that correctly stated that aerosol-like particulate pollution has a cooling effect, while greenhouses gases have a warming effect. In fact, in the Radiative Forcing Components table above, issued by the IPCC, we can see that aerosol emissions do have an effect that counters the prevailing warming trend also according to 21st-century science. However, the article by Rasool and Schneider seriously underestimated the effect of greenhouse gases on climate; therefore, it depicted the twofold pollution-caused radiative forcing as having a prevailing cooling effect instead of a warming one. Science is not about prophecies; this example just shows the evolution of the research on the subject. Strange as it may sound, the opposite conclusion is just the result of a finer tuning of the same basic intuitions of that obsolete article. Do we oppose the fundamental scientific idea that concrete quantitative considerations do have an effect on the results of a theory? Truth is always concrete, as Lenin said.
Secondly, there never was a "scientific article of faith" proclaiming a coming ice age. Perhaps the mass media did promote ice age hysteria, but this was never done with the support of complacent scientists.
Somebody actually counted how many scientific articles (peer-reviewed, published in scientific reviews etc.) during the last few decades predicted global warming, how many predicted global cooling, and how many took a neutral stance. One of such quantitative researches  gave the following graphical result:
Figure 8. The number of papers classified as predicting, implying or providing supporting evidence for future global cooling, warming and neutral categories as defined in the text and listed in Table1. During the period 1965 through 1979, our literature survey found 7 cooling papers, 19 neutral and 42 warming. In no year were there more global cooling papers than global warming.
In striking contrast to the correct opinion already building up in the scientific community during the late Seventies, a survey by Maxwell T. & Jules M. Boykoff showed a different picture in 4 influential US newspapers as late as in the 1988-2002 period. The Boykoffs categorised 636 articles published by The New York Times, The Washington Post, The Wall Street Journal, and the Los Angeles Times. Notwithstanding a scientific consensus had already been reached on the issue, 52.7% of the articles paid the same attention to global-warming and "sceptic" positions. Only 5.9% of the articles explained global warming without counterbalancing it with an opposing position - actually less than the number of articles that provided only "alternative" views.
The funny part of Baker's position is that he apparently believes that scientists are more politically motivated in their alleged pro-capitalist actions than journalists and conservative politicians. Isn't it an astonishing position for a Marxist? The true story is that public acceptance of global warming as a fact was consciously delayed by the very same lobby network that is now spreading scepticism.
Argument 5: A computer model is used (as for everything else)
Baker devotes a sizeable part of his article to criticise the use of computer models. He lists all approximations used in computer modelling and also gives us some insight into the problems of his own job.
What does such an argument imply? If one has to take it at face value, we should be proposing to abolish physical models. Whoever, like Baker himself, has worked with models knows that they can only roughly represent a partial aspect of reality; also the model of the most elementary physical processes are filled with instrumental data (underived by general principles), linearisations, simplifications, omission of "secondary" components. This is a great confirmation of the dialectical position on gnoseology, which denies the possibility of final knowledge and conceives the process of knowledge as a never-ending approach to an infinitely complex objective reality.
As Lenin wrote in Materialism And Empirio-Criticism:
«Two questions are obviously confused here: 1) Is there such a thing as objective truth, that is, can human ideas have a content that does not depend on a subject, that does not depend either on a human being, or on humanity? 2) If so, can human ideas, which give expression to objective truth, express it all at one time, as a whole, unconditionally, absolutely, or only approximately, relatively?»
The partiality of any knowledge does not necessarily imply partisanship in a bad sense. Here, the detail that computers are used is completely irrelevant. Models are an important part of science, and computers have just helped us in the construction of more complex models that we could never solve by hand.
Bourgeois theoreticians, with their open contempt of broad generalisations, have always had an evident hostility towards general models, usually favouring narrower correlations. In fact, this argument has been often used against Marxist economics and sociology.
What is the alternative proposed to models? Just statistical correlations, like the ones Baker likes so much: those produced by Svenmark and other deniers to suggest that the sun's magnetic field is the only factor in climate change. These studies just superimpose two graphs and try to demonstrate something in that naïve manner. RealClimate effectively mocked them using the following diagram  that shows a mysterious correlation between Republicans in the US Senate and sunspots (the mock correlation undergoes an unexplained inversion event - a "phase shift" - around 1986):
The point is that correlation is not causation. Something can always happen together with something else and yet not be its cause: it could be its effect, instead, or both events might be caused by a third cause. A typical example to mock correlationists is to underline how global average temperature has increased as the same time as Caribbean pirates have decreased.
This logical fallacy is called cum hoc ergo propter hoc and is a typical feature of the pigeon brain. If a pigeon is fed while it is performing, for instance, a certain neck gesture, it will "infer" that the gesture caused the food to be supplied and will repeat the gesture in an attempt to obtain more food. For this reason, pigeons are sometimes described by ethologists as exhibiting a "superstitious behaviour". 
Models give more insight (and make us cleverer than pigeons) because they explore the actual causal connections between processes.
[To be continued...]