Posted on February 27th, 2014
“Science” is one of those words with many meanings, some of which conflict with others.
In a psychology course, once, I was told that psychology was a science, and that the goals of psychology are “understanding, prediction, and control”. The professor went on to explain that prediction provides proof of understanding: if we can perfectly predict something, then surely we must understand it. And that control is the ultimate means of ensuring accurate prediction. This took an entire class hour to explain, because I was a mere student in Psych 101. At the time, I did not feel like I controlled anything, I liked prediction, and I loved the feeling of understanding. So I thought maybe I should be a psychologist. This was early in the course, and I later learned that I suffered from every disorder psychologists had ever named, and so needed a psychologist.
I learned that psychologists conduct experiments, the way crop research was once done, randomly assigning subjects to “conditions”, and later measuring the differences between the groups. When statistically significant differences are found, the conditions are assumed to be the cause, and a causal relationship is established.
I thought about astronomers, who are not very good about controlling the course of planets, and of weather forecasters, who don’t seem to be able to control the weather.
Astronomers and meteorologists think of themselves as scientists, so perhaps neither prediction nor control are critical aspects of a definition. Maybe understanding is all we are after.
For much of science, “statistical inference” is a critical process. Sample size must be large enough, assignment to condition must be random, only statistical differences in the outcomes counts for anything. Of course, while the mathematics understood by astronomers and meteorologists would make any psychologist’s head spin, it doesn’t look like the statistical methods of psychology — t-test, analysis of variance, multivariate analysis — have anything to do with their world.
I was trained as a research psychologist, and for years embraced my training. But ever since, I’ve wondered if there isn’t another reasonable way to do science.
Consider geese coming in for a landing on the water. Casual observation reveals that some flop from side to side, a move called whiffling. But why do some do this and not others? A bit more observation reveals that only the large birds, with the biggest wingspans, seem to do it. And then we reason: if a flock is to land together, and some have longer glide ratios than others, some will be able to coast to a landing, and others will have to brake. Or some will brake, and others will have to brake harder. Spilling the air by whiffling breaks the lift generated by the wings, and allows a faster decent. For the fastest decent, it seems, some geese have been observed coasting upside down.1.
Our speculation continues. What do we know about the largest birds? If we’ve gone this far, we might as well keep watching. Soon we learn that it is adult male geese who are much larger than the ladies, who have larger wings and must have more lift. And while we watch this again and again, to be sure, we discover something else: the big boys follow their wives. And so it must be them who do most of the whiffling. We note that the honks of the two while flying are considerably different. We note that he rarely objects to a direction she chooses. We begin to suspect that he loves her, and that someone’s heart will be broken if a hunter takes one out.
And then we get thinking about that V-formation pap we were handed in elementary school: that geese fly in a V because of the aerodynamics, that it makes flying less effortful. But surely this can’t be true, we reason: Are geese the smartest of birds, and others didn’t think of this? That pelicans merely have to plod along single file? Or could there be some other cause, like a desire to keep your beak out of someone’s butt? If we look at fighter pilots, we note that they take the same V-formation when flying over the stadium. And we discover that not only are they lateral and behind or ahead of adjacent planes, but they are also above or below.
And so we look to our geese again. If it is a desire to stay out of a goose’s butt that shapes the V, surely individuals in a flock must fly at slightly different altitudes. Let Google prove this: search for “canada goose flight”, then look at images. Find photos taken from the side, rather than from below. You’ll see that there are no photos in which all geese are on the same plane.
While we ponder this, we get thinking about reaction time in geese. If the goose ahead of you turns into your path, or begins to brake, you can’t even know this until you see them. And with your eyes on the side of your head, you may not see them immediately. So the details of the V-formation are surely controlled, to some extent, by the need to keep other birds in your field of vision.
If you did all this thinking and observation as I’ve done, and reached similar conclusions, you’d likely have a comfortable feeling of understanding, and modest skill in predicting. You could predict when whiffling would occur, the positions in the flock that would be most likely to do it, and the sex of the whifflers. You’d get it right. But that wouldn’t be good science. Scientists would need to study dozens of species to see which whiffle. They’d need to weigh birds before takeoff, and attach cameras to their backs. They might need to shoot them all after their experiment and autopsy them, to see if they could be the first to find a whiffling organ. And it wouldn’t be science until their learned findings were published, in unreadable fashion, in an expensive refereed journal that no one will read.
I think that careful observation, clear thinking, and thoughtful inference can be good enough. I think that studying whiffling all my life is not a good idea, because there are so many other things I want to know. The methods of traditional professional science are intolerable. The characters in that bad play chew up your tax dollars. Have no pressure to explain to you what they learn. Find joy in doubting everything, and are never quite certain of anything. Traditional professional science in most fields2 is never better than the alternatives, though those professionals think they are better, and must be more important because they are paid to treat us as lessers.
The problems of science were inevitable.
- To know much about something, we must focus, and set aside other interests as we focus. So as it developed, science learned more and more about less and less. It didn’t take too long in this procession for scientists to shift from studies of everything to studies of almost nothing. In the past decades, we’ve shot past the goal of understanding everything.
- To earn a living doing science, we needed to turn it into a profession, with a peer recognition system, norms for obtaining recognition, and accepted procedures for everything. In the process, we both impressed and alienated mortals, who were not wise to our ways.
- To generate more work, we needed to doubt everything (except our own results). The guaranteed last sentence of every published study seems to be “Further research is required.” A problem solved is a research specialty lost.
- Since the public would not support us if we were to proclaim our findings on street corners, we had to get someone else to fund this: either the public (through taxes and grants) or the public (through products purchased from the companies for whom we worked) or the public (through sending students to our classes to hear disinterested derivative reports on what was happening in our general field some years ago.) Getting the public to pay directly for what we know — perhaps by being paid by Psychology Today or Popular Science for a submitted article — is difficult: we never learned to write in a way that was readable by mortals. And an entire career’s learning can often be summarized in a few column inches.
For now, I’m willing to call the methods of traditional professional science Rotten Science. Of course, those individuals who consider themselves traditional professional scientists are not rotten. They’d likely be as intrigued by whiffling as you or I. I think they can be taunted or maybe recruited by some of the alternatives I propose below.
Other Scientific Methods
What would a better science look like? Let’s imagine what rules it must adhere to, before we imagine its methods. Good science should be responsive, answering questions that someone is asking. The more wanting an answer to a question, the more responsive it can be. Good science should consider costs and benefits. What does it cost to come up with an answer? What benefit does it convey? Our whiffling study cost nothing, certainly an amount less than its slight benefit. Good science should be able to live with uncertainty. We don’t have time or resources to answer every question. As a result, “I don’t know” is usually a perfectly good answer to a question. “We’re working on it” is not as good. Good science should be something that everyone can do. Good science should identify some more relaxed rules for inference. In some cases, we should be able to infer something from a sample of one observation. For instance, if you just watched a goose flying upside down, you “know” these things already:
- Geese are generally able to rotate their heads 180 degrees in either direction. In both the still photo sequence and the video, that’s what two different species of geese did. Two observations is not much, says the scientist. But we don’t go far when we extrapolate from “of 2 geese flying upside down, both turned their heads to remain right side up” to “when flying upside down, a goose turns its head to remain right side up.” Of course, we might one day find an exception to this. At that time, I’m comfortable revising my conclusion to “when flying upside down, a goose [is likely to] turn its head to remain right side up.” Given many more such contrary observations, I’d be willing to say “when flying upside down, a geese [sometimes] turn turn their head to remain right side up.”
- Some geese are able to fly upside down. Of course, if we wanted to praise these birds, we could say “Geese can fly upside down.” If we wanted to diminish them, we might say “There have actually been 2 instances of geese flying upside down briefly.”]
- We are able to impute motive to a goose flying upside down: to descend rapidly. We get to this position by extrapolating from whiffling, which seems to only be done by the bigger birds when the flock is landing — birds with a longer glide and more in need of braking — and extending to the realization that if wings in the normal orientation cause lift, those that are inverted will cause descent.
Good science should seek multiple ways of confirming or qualifying our previously conclusions. Now that we wonder about 180 degree rotation in flight, can we ever see this in other cases? Yes, all birds, it seems, can groom the feathers on their spines, a 180 degree neck rotation. Can other birds fly upside down? Google Images will answer this question for you. If you search for “bird upside down flight” you will see Snow Geese, Bald Eagles, Gulls, hummingbirds, and an Elegant Tern all flying feet up. Can a bird fly right-side up, with its head upside down? Again, Google images will show you a White Ibis with its head rotated, preening a feather in flight. And how do we know that an inverted wing would pull a bird or airplane down, rather than up? Try flying an airplane upside down, without changing the angle of the flaps. [No, don't try this at home.] For methods, I want to focus on animal behavior as an area of interest. What is it that animals do? How do they do it? What seems to trigger those behaviors? What behavioral similarities are there across species?
Good science is not a method, but an approach to life. We can’t say that science starts with a hypothesis, like they told us in high school, because there would be no hypothesis without many observations and inferences. There is no evident beginning of good science. But there are some principles that all good science must embody: observation, inference, faith, and suspicion. As we think about our observations, we will likely form and reject hypotheses.
There are many ways to observe animal behavior:
- surreptitiously observing the animal in the wild, with binoculars, telescope, microphone, tracking collar, or other means.
- raising a domesticated or wild animal in captivity, allowing it to imprint, and keeping it captive
- capturing a wild animal, making it captive, and studying it while a captive
- raising a wild animal for an interval in captivity, allowing it to imprint, but then as soon as the animal reaches the appropriate life stage, releasing it and accompanying it where it goes
- taming a wild animal, and allowing it to remain wild.
- reading about any of the above, watching a video, or otherwise vicariously observing an animal.
Most readers will have a preference for one of these methods of observation, and will have strong opinions that their method is best. In the end, it may not matter what method we use, if we use it well. But I believe that all methods, in combination, provide the best opportunity to combine observations and to generalize in our inferences.
For example, any horse owner knows that a horse pins its ears back when its angry. Any dog owner knows that dogs do this. And readers of “Among the Bears” (see Suggested Readings) will know that bears do this. It would not be hard to infer from these three sets of observations that animals pin their ears back when angry. Reflection takes us further. Birds don’t pin their ears back ever, because they don’t have pinna3, or external ear flaps. Snakes don’t either. So our conclusion becomes a bit more cumbersome: Animals with pinna may pin them back when angry.
I also believe that every one of the observation methods listed above can normally be improved. Consider the bird watcher, using method 1. Most birds are heard before they are seen, and while most birds can be heard, few will be seen at all, even by a reasonably patient observer. There are alternatives to missing all of the action. Bird watchers could more regularly carry directional microphones connected to earphones (an auditory equivalent of the spotting scope) and to a recorder (an equivalent to the camera.) They could improve their chances of seeing more by being seen less: bringing a comfy folding chair and a portable hide or blind4. Tradition seems to have defined the methods of birders as firmly as it defines the methods of traditional, formal science.
Each observation method can provide a vast body of information about the critter. But it is not until we have used two or more observation methods, and observed two or more species, that we move toward a trustworthy position for broader inference. For instance, Joe Hutto discovered that other wild animals seemed to lose their fear of him when they found him in the midst of (normally very timid) wild turkeys. And Ben Kilham reports that wild bears are much more comfortable approaching when they find him socializing with other bears (that he has fostered.) I have found that I can squeeze as many as 20 raccoons onto my patio when I’m hand feeding one or more of my raccoon buddies. These three reports come from two different methods: method 4 for Hutto and Kilham, and method 5 for myself. They come from three different species. And so it might be safe to infer that [some] wild animals will [sometimes] sense the emotional state of other wild animals of the same species (raccoons and bears) or very different species (turkeys), and assume that same state. The feeding of the raccoons provides assurance that their is food and that it is safe to come and get it. Sitting against a tree with a turkey can relax nearby deer and squirrels; sitting with bear as a companion can relax an approaching bear. Each observation method can serve to validate or refute a conclusion from some other method. Each species observed can help us extend or narrow a generalization.
How We Are Better
It seems as if most people want to believe that humans are better than everything else. (Many also believe that their country, and their home town, are best.) I don’t share their insecurity. See my rant here.
I think that the question of how we are better is ignorant. We aren’t better at what we do than other animals are better at what they do. We don’t have any special characteristics that is not share, to some extent, by some other species – except perhaps for the notion itself that we are better. Other animals seem to have more modesty.
Believing we are better is the foundation for all sorts of prejudice and trouble. In our science, we do a disservice to those at both ends of the telescope or microscope when we think this.
At this point, I am sure, I have no readers surviving. Its one thing to attack scientists. Its another to attack my entire rotten species.
For the reader who decided to keep going… My first draft of exploring this assertion was simply to wonder how humans are different. That turned out to be part of another question: how are species the same? The similarities and differences between species can inspire and correct inference and help us discover the appropriate bounds of extrapolation. Since so many of our books on nature take the condescending approach of author as better than subject, and difference as proof of this, I prefer to focus on similarities, and usually like to leave our species out of my thinking. The world of bears, turkeys, and raccoons is large enough for me.
Couturier, Lisa. The Hopes of Snakes and Other Tales from the Urban Landscape. Beacon Press. January 2, 2005. 150 pages.
Lisa Couturier writes with great power and sensitivity, pulling the reader in, teaching and thrilling and spinning a yarn like no other nature writer. Along the way, the words tumble together in new ways, and the charm and delight flows in a torrent. In the end, we are moved, we have learned, and we want more. I hope you will share my delight with this book, and share it with friends. I’ve bought about 30 copies that I’ve given away to people I knew would enjoy it.
Hutto, Joe. Illumination in the Flatwoods: A Season with the Wild Turkey. Lyons Press. October 31, 2006. 256 pp.
Also see the video, My Life as a Turkey which PBS derived from this book: http://video.pbs.org/video/2168110328/. The video provides less illumination, but more intense visualization.
Kilham, Benjamin. Among the Bears: Raising Orphaned Cubs in the Wild. Holt Paperbacks. March 1, 2003. 304 pp.
Most of us will never get to have bears in the basement, or to walk in the woods with bear cubs, or to stick our heads into a bear’s den to see how things are going. So if our only bear contact has been through TV, this book will be valuable.
Kilham writes well enough that he largely disappears from the story, and you can imagine doing what he is doing, learning what he is learning, walking in his shoes. And in so walking, we learn how little we previously knew about bears, become fascinated with their lives and society, and want to know more.
Much of the power of the book comes from stories. Stories of how a mother bear introduced her cubs to the author, of how she protected the author from a big male bear that saw him as a rival, of how bears form friendships and share food with their friends, how bear cubs love to play, how mother bears go about teaching the kids, how bears can love a man, and show that love.
As a story teller, Kilham is terrific. For spending every daylight hour outdoors in New England, summer and winter, he never seems to get hot or cold, never seems to get wet when it rains, never gets tired, never spots a black fly or mosquito.The story is not about him. It is about the bears. They couldn’t have asked for a better translator or advocate.
I don’t think this book delivers as much insight into ourselves as “Illumination in the Flatwoods: A Season with the Wild Turkey”, which is slightly better written, but the two authors approached their subject matter in the same way, with similar results: they devoted all of their time to raising young wild animals and re-introducing them to the wild, and along the way were admitted to a secret society.
I bet you’ll give it 5 stars too, and go on to buy his other book – “Out on a Limb: What Black Bears Have Taught Me about Intelligence and Intuition” – which should be read after this. And if you haven’t read “Illuminaton in the Flatwoods”, better hurry up and do that, too.
- For a series of photos showing a greylag goose whiffling into an upside down position, see http://www.dailymail.co.uk/news/article-1184566/Whiffle-wind-Pictures-flying-gooses-incredible-contortion-slow-down.html For a video of a Canada goose rolling upside down trying to brake, see http://www.newscientist.com/blogs/nstv/2012/01/first-slow-mo-video-of-goose-flying-upside-down.html Notice how fast the goose drops when its wings are pulling it down, rather than up! ↩
- Drug researchers are surely scientists, and surely do useful things for the public. Scientists in “applied” careers — museum curator, guide, teacher — are surely useful. My beef is with others. ↩
- see http://en.wikipedia.org/wiki/Pinna_(anatomy) ↩
- see http://www.gowildlifewatching.co.uk/Wildlife%20Watching%20Hides.html for examples. ↩