Environmentalism, to most people, means protecting the Earth's pristine ecosystems from the defiling impacts of humans. The problem is that these ecosystems never existed in a stable, "pristine" state.
A recent article in Science discusses this in addition to the difficulty of even knowing what these ecosystems "originally" looked like.
From a North American perspective, ecologists have traditionally looked to the writings of European pioneers in order to ascertain the original state of the continent's ecosystems. We've learned since that the Native Americans heavily manipulated these ecosystems (while living in surprisingly large, organized civilizations). One of their most dramatic tools was the prescribed burn. In the wooded Mid-Atlantic and Northeast, these burns selected for fire-resistant species, most notably nut trees such as oaks, which produced food both directly for humans, and indirectly, by supporting game species. As we speak, these "natural" oak forests are increasingly being overtaken by fire-sensitive species (e.g. maples) that are more tolerant of shade and compete better within intact forests - which more likely represents the state of these forests before humans.
Similarly, concern arose in Yosemite National Park recently as trees began to die of fungal root rots - often collapsing onto buildings in the process. These dense forests hardly resembled those encountered by European travelers over a century earlier - when an oxcart could easily be driven between trees. Fire was again the explanation. With this human-selective pressure removed, the forests reverted to a common Western conifer disease cycle. Shade-intolerant tree species are increasingly out-competed by shade-tolerant species (that are susceptible to fungi). When enough of these shade-tolerant, but susceptible trees accumulate, root rot epidemics take hold and expand each year - leaving open land for the shade-intolerant species to recolonize.
Even when paleontological techniques have been applied to identify ecosystem parameters before any humans were present, the picture remains fuzzy. Currently, North American forests are known for existing in certain combinations of key species (e.g. oak-hickory, oak-hickory-pine, beech-maple and maple-basswood). When ancient deposits of pollen were analyzed, it was found that many of these ecosystems had different combinations of dominant trees in the past. We're still living in a period of oscillating ice ages and interglacial periods. North American ecosystems retreated south each time glaciers advanced and expanded north again as temperatures increased. Because each species has unique tolerances of not only temperature, but also soil type and rainfall, they all shuffled and re-assorted with the changing climate.
The author specifically points to the Big Woods of Minnesota, which apparently formed on the site of a great savanna-prairie around 1300 C.E. as successive years of drought reduced fire fuel load, allowing trees to invade the grasslands. Ironically, these forests are now being changed again as earthworms are re-invading (very slowly!) since the glaciers last wiped them out. Many of these "invasive" earthworms were formerly native!
I think environmentalists who are too hung up on the "restoration" of ecosystems do a disservice to conservation as a whole. It's important to understand how our activities impact the Earth's ecosystems, but not all these changes are necessarily negative. One day we'll all be gone and the Earth will evolve completely new organisms and ecosystems just fine without us - so there's no moral argument to "protect" ecosystems. The environment is only valuable to the extent that humans assign it value. I propose we take a practical approach:
1) Maximize the environment's ecosystem services
e.g. invasive zebra mussels are cleaning up Lake Michigan
2) Work to restore culturally-important ecosystems
e.g. breeding blight-resistant chestnut trees
3) Stop worrying so much
It just prevents the average citizen from taking action