"Prudens quaestio dimidium scientiae" ("Half of science is putting forth the right questions" or more simply "To know what to ask is already to know half") -- Roger Bacon
"Argumentation cannot suffice for the discovery of new work, since the subtlety of Nature is greater many times than the subtlety of argument" -- Francis Bacon, Aphorism XXIV, Novum Organum (1620)
"All knowledge that is not the real product of observation, or of consequences deduced from observation, is entirely groundless and illusory." -- Jean-Baptiste Pierre Antoine de Monet, Chevalier de Lamarck, System analytique des connaissances positive de l'homme (1830: published posthumously)
There is nothing like geology; the pleasure of the first day's partridge shooting or first day's hunting cannot be compared to finding a fine group of fossil bones, which tell their story of former times with almost a living tongue. -- Charles Darwin, letter to his sister Catherine, 1834
"Science is what we do to keep from lying to ourselves." -- physicist Richard Feynman
"Science works with testable proposals. If, after much compilation and scrutiny of data, new information continues to affirm a hypothesis, we may accept it provisionally and gain confidence as further evidence mounts. We can never be sure that a hypothesis is right, though we may be able to show with confidence that it is wrong. The best scientific hypotheses are also generous and expansive: they suggest extensions and implications that enlighten related, even far distant, subjects." -- paleontologist Stephan Jay Gould, "Sex, Drugs, Disasters, and the Extinction of Dinosaurs" (Discover Magazine, 1984)
BIG QUESTIONS: How do we use the fossil record to understand the biology and evolution of extinct organisms?
The Hypothetico-Deductive Method
Science is not simply "a body of knowledge". Rather, it is the systematic acquisition and application of knowledge about the structure and behavior of the physical universe gained via empirical evidence through observation, measurement, and experimentation. It can be described as a type of inquiry into nature characterized by the availability of empirically testable hypotheses.
Science starts from several important observations about the world around us:
Science is thus a self-correcting mechanism: it contains within its operation the means to get rid of old, less-accurate models and replace them with more accurate ones.
Argument: a connected series of statements intended to establish a proposition, consisting of one or more premises which support the conclusion. (In the case of Science, the premises should be a series of hypotheses tested against observations.) Many, many fields fields of intellectual endeavor (politics, philosophy, marketing, etc.) relay on arguments. However, Science is distinguished (in a subset of fields: we can add some others like history, economics, etc. in here) by relying on not merely argument butindependent evidence: in other words, it is empirical.
There is much talk about the "Scientific Method", often characterized as a hierarchy of observations --> hypotheses --> theories --> laws. This is in some ways an oversimplification, and in others misses the point. These are more the products of Science, not their method. Perhaps a better descriptor of the "Scientific Method" is simply: Physical Evidence and Reasoned Logic (PEARL).
That physical evidence goes by several names: observations, data, measurements, all meaning the same thing: qualitative or (more commonly) quantitative attribute of a phenomenon. In principle, different observes should be able to make the same measurements/descriptions and find the same value.
Of course, we face the issue of uncertainty: there are limitations of observations (for example, the accuracy and resolution of instruments) so we will not always find the same exact value. Additionally, there are probabilistic aspects to the nature of the Universe, especially at the quantum level.
Raw observations are good, but we need to do more with them. In other words, there must be some form of data analysis. The observations are compared to each other in some fashion (normally some form of mathematical or statistical plot) in order to discover potential patterns, and from that to test different hypotheses.
Collectively, raw observations and analytical results are what we call evidence. We use this evidence to infer what is happening; that is, evidence is used to get to conclusions.
After observing some phenomenon, a pattern often emerges. We can state this pattern formally as an hypothesis. Thus, contrary to its colloquial use, an hypothesis is NOT an "educated guess", but rather "a formal statement of a pattern that appears to exist in a set of observations." In contrast to theories, hypotheses are primarily about the pattern itself, not about an explanation of the pattern.
Unfortunately, humans cannot help but see patterns: castles in the clouds, faces in random objects, "lucky streaks" in games, etc. Not all perceived patterns are real! So to evaluate whether a pattern that we perceive (an hypothesis) might be true, we must test it. We refer to this as "submitting it to falsification": subjecting the hypothesis to some evaluation where it could in principle be shown to be incorrect. Not all hypotheses are falsifiable (synonym: testable): some are purely subjective (e.g., "chocolate is better than vanilla") or involve metaphysical qualities or entities which cannot be measured (e.g., "true justice is superior to true wisdom"). We might hold these as important concepts, but they are outside Science. Indeed, to paraphrase the late philosopher Christopher Hitchens, "What can be asserted without evidence can be dismissed without evidence".
Other hypotheses can be potentially falsifiable with total knowledge of the universe, but we are currently [and perhaps forever] incapable of evaluating (e.g., "the flesh of eurypterids ('water scorpions', extinct for 252 million years) is an effective cure for athletes foot"). Ideas of this second sort are speculations: nothing wrong with them as such, but they aren't particularly useful.
Even good scientific hypotheses remain mere speculations until submitted to a test (often called an experiment). An experiment (i.e., a test of falsification) must be constructed in such a way that the hypothesis could yield observations that demonstrate that the hypothesis is false. For example, we can speculate that "my horse can outrun any horse in the world." However, until we gather evidence to test it, we won't know if this is actually the case (no matter how much we want to believe it.) A rather simple experiment for this: a horse race. If another horse outraces it, then the hypothesis is falsified.
This procedure is called the "hypothetico-deductive method", and is basically what experiments in Science are all about. To put it in its basic form "If you were wrong, how would you know it?"
Note that part of developing a good experimental design is actually phrasing your hypothesis properly. For example, we might be interested in the presence of fossils of ceratopsids (horned dinosaurs, like Triceratops) in rocks the last 15 million years of the Cretaceous Period in continental Africa. If we state our hypothesis "there were latest Cretaceous African ceratopsids", than how many observations do we need to make to demonstrate this is true? Does a single observation with negative results show us there is no ceratopsids in latest Cretaceous Africa? No. How about ten negative results? A thousand? Until we have excavated all latest Cretaceous rocks from Africa, we cannot demonstrate that there are no ceratopsids. However, by stating the hypothesis as "there were no ceratopsids in latest Cretaceous Africa", then a single positive observation of a ceratopsian is all we need to confidently overturn (falsify) this hypothesis.
The above is an example of employing a null hypothesis. The null represents a form of the hypothesis that must be rejected before we even accept a phenomenon exists. If we cannot reject the null hypothesis, there is no reason to think that the phenomenon in question is worth considering. If, instead, we find we can reject the null (in the instance above, finding a latest Cretaceous ceratopsid in Africa), than we can go on and examine the situation in more detail.
The hypothetico-deductive method shows that we can confidently reject hypotheses, but that we cannot "prove" them in an absolute philosophical sense. As the number of observations and experiments which fail to reject an hypothesis increases, we can be more and more confident in its truth. However, ideas in Science are only provisionally accepted: that is, we often use statistical measures of confidence (plus/minus readings, error bars, and other demonstrators of degree of support). Uncertainty is a staple part of Science. However, some ideas are so overwhelmingly well-supported that to reject them at present is perverse: these are what we call "facts". (Similarly - using the example above - if we made many excavations in latest Cretaceous continental African rocks and continuously failed to uncover ceratopsid fossils but find plenty of other dinosaurs, we will provisionally reject the idea of African horned dinosaurs, but recognize that a single fossil could overturn this rejection.)
The following (from Thomas Kida's Don't Believe Everything You Think) are a useful set of characteristics for thinking like a scientist:
Something worth noting concerning the issue of paleontology, historical geology, evolutionary biology, astronomy, and other sciences which are concerned primarily with events which have already happened: i.e., historical sciences. These are concerned with actual particular events and cases just as much as general patterns (as opposed to, say, particle physics or organic chemistry, where all collisions of the same particles or folding of the same protiens are identical). As a consequence, although we can perform some kinds of experiments or observations under controlled circumstances to duplicate the past conditions, in general we are looking at evidence of the past event.
That doesn't make it any less scientific, nor does it dismiss the ability to do repeatable observations. "Repeatability" in this case is where different observers--or the same observer in multiple different examinations--can make the same observations of the same set of data. (For example, if a scientist asserts that a particular layer of clay shows an enriched abundance of the metal iridium, consistent with an asteroid impact, other scientists can sample the same layer and look for this material, and the original research can resample the material to test that it is there.)
There is a cliché that paleontologists are a kind of "detective", and that isn't a bad comparison. Much like forensics experts in crime scene investigations or medical examiners, we examine a series of data to develop a hypothesis (or multiple hypotheses) to explain the observations at hand, and test these hypotheses against both the currently-known data and additional relevant data we look for.
In general, it is best to follow Carl Sagan's "Baloney Detection Kit":
However, just because something is a theory doesn't mean it is an accurate map of how the Universe really operates. A scientific theory is more specifically a comprehensive framework for describing, explaining, and making falsifiable predictions about related sets of phenomena based on rigorous observation, experimentation, and logic. Scientific theories are not necessarily correct (that is, accurate maps of the Universe), but before we accept them they should at least match our current observations.
A key element of both theories and hypotheses is that they are predictive: that is, by using them you can make estimates before hand about observations not yet done. In the case of historical sciences like paleontology and geology, "predictions" don't necessarily mean "what things will be like in the future", but rather "what will we be likely to discover about a past event, which we have not yet examined".
Contrary to some statements (even by reputable scientific organizations), scientific theories are not limited to describing only large scale ("universal") phenomena nor ongoing phenomena. Thus, "one time" only events (such as the impact theory of the extinction at the end of the Cretaceous or the Big Bang at the beginning of spacetime) or phenomena limited in scope (such as the theory of thermohaline circulation as the major driver of modern climate, or again of the impact cause of the Cretaceous extinction) are indeed subject to theory, and these theories can be explored and tested via parsimony, fecundity, consilience, auxilary hypotheses, and so forth. Nevertheless, the most important (and most fecund) theories ARE about large-scale ongoing phenomena: atomic theory of matter; periodic theory of the elements; special relativity; the germ theory of disease; the theory of evolution by means of natural selection; plate tectonic theory of geology; and so on.
It is very common to find that in certain spheres of Science there are multiple different (and sometimes mutually exclusive) theories proposed for the same observations. Indeed, this is what a lot of scientific research is about: the creation of and testing of new theories and their auxiliary hypotheses. Hence there are fields like theoretical physics in which new models of the operation of the universe and its various components are proposed and assembled based on previous observations, logic, conjecture, and speculation. Other scientists (experimental physicists, observational astronomers, etc.) themselves look for observations that could in principle reject some or all of the components of these theories. This research involves creating sets of experiments whose results will be different under different alternative theories about the phenomenon involved. We can test and reject theories in part in the same fashion as we do hypotheses: that is, by parsimony, consilience, and by tests of falsification.
One last comment: what are scientific laws? Popular accounts of the scientific method suggest a hierarchy of observation -> hypothesis -> theory -> law, but this is not correct. The phrase "scientific law" in the Sciences was largely been abandoned in the 20th Century. Many of the traditional "scientific laws" were simply scientific theories that can be rendered as mathematical equations. As a consequence they tend to deal with relatively simple and more easily measurable phenomena. "Scientific laws" were thus no better nor worse than other scientific theories at withstanding rejection: for example, Bode's law of planetary orbital distance wound up being a coincidence more than a law; Newton's laws of motions only apply to certain gravitational conditions and speeds; and so forth. "Scientific laws" can be useful in some circumstances (e.g., calculating gas pressure, volume, temperature, or number of particles given the other variables using the ideal gas law PV = nRT), but there are many realms of Science where the phenomena are too complicated to be reduced at present to law-like forms. This holds true of much of geology, climatology, biology, and especially behavioral sciences. So be very careful if you hear from someone who proposes some "law" of biology or paleontology or anthropology or psychology!
There is a subset of theories that have withstood substantial repeated tests and modifications and survived if not unscathed, at least strongly supported and the victor against all challengers so far. These incude (but aren't limited to) the atomic theory of matter; the theory of evolution by means of Natural Selection; plate tectonics theory; the germ theory of disease; etc. We honestly don't have a good term to distinguish these theories from the more run-of-the-mill, still-in-play types. (One might suggest facts, and there is some merit there, but a "fact" might be a better synonym for a well-substantiated observation rather than an entire theory.) The British evolutionary biologist Richard Dawkins proposed (in 2009) using the new word theorum for this class of theories. Time will tell if this nomenclature will catch on.
Regardless of what we call them, we can describe this collection ideas as theories which are to the best of our knowledge "true", but this requires the caveat that absolute "Truth" is empirically impossible to find (although it would be extraordinarily perverse to reject these things as being "true" in a general sense without some extraordinary new evidence to the contrary.) These winning theories can be thought of as the all-time prize-winning champions in their respective fields, having faced and defeated all challengers. In principle (indeed, "in theory"! :-) they could be defeated by a newcomer, but the weight of evidence so far is with the champions.
On the other hand, many scientific questions (especially technical and precisely phrased ones concerning matters which we can have access to the appropriate observation) have correct answers: that is, they are not matters of opinion (as are questions in a great many of other fields).
Not all viewpoints are equally valid in Science:
Concepts within science are subject to change with new discoveries. That is in fact what scientists DO for a living: make new discoveries!!
Often, people ask "what do scientists believe." This is the wrong question! Science is not about belief; it is about discoveries and about the methods by which those discoveries were made and tested.
Through Science, we have discovered many aspects of nature. Here are some of the largest level aspects (finer details would be those covered by different content disciplines):
Why write & publish papers?
Scientific papers might be very short (1-2 pages), or monographs (dozens to hundreds of pages long), but most are typically between 4 and a few dozen pages long. Monographs are primarily used to thoroughly document a single particular topic: a complete description of the anatomy and biology of a particular species or group of species; a review of the geography or the geology of a particular region; the results of a particular space probe; etc.: in other words, topics with a large amount of observations in a very narrow topic. Very short papers tend to simply announce a new discovery, document an important new observation, or respond to a particular criticism of previous work. The middle range papers are where most the hypothesis testing goes on.
It was once very common to have single-authored papers, or maybe just two authors. These have become fairly rare, and it is common to find papers with a half-dozen or more authors. In fact, in some cases there can be dozens of contributing authors!
How are scientific papers assembled and published? In general, they follow the pattern here:
Paleontology and forensic science (e.g., "crime scene investigation") have a lot of similarities. Both involve reconstructing past events from the limited evidence (often circumstantial) to find one (or often several) likely explanations. This is just as true for trying to reconstruct how ancient organisms lived: that is, paleobiology. Paleontology has many different uses: reconstructing past environments, past events, sequence in geologic time, etc. But for many people, paleobiology (i.e., "bringing fossils to life") is the main draw.
So how do we go about this?
Step 1: Identify the Parts: Determining what particular organs, structures, etc. you are seeing in a fossil can be complicated. In some cases, they are animals or plants which are no longer extant, and may have parts that do not correspond exactly in size and shape to the homologous parts in living relatives. And in some cases, they may even have parts that are NOT homologous to the parts of any living organism! Hence, the need to use comparative anatomy to establish what parts you see in the fossils.
Step 2: Classify the Organism: Once you know the parts, you can compare it to already-known specimens. It might very closely match the anatomy of species previously established from fossils; in this case, the fossil is most likely a new specimen of that previously-named species. Or it might differ to some degree from the known fossils: in this case, the paleontologist might make the case that fossil belongs to a new species, a new genus, or indeed some other previously unknown branch of the Tree of Life.
Step 3: Restore the Missing Parts (including Soft Tissue): No fossil will be entirely complete: after all, most are missing soft parts, and even exceptionally good fossils are missing significant portions of the hard parts. So we need to fill in the gaps. For the hard parts this can often be done by mirror imaging the equivalent part on the other side of the body (for a bilaterally-symmetrical organism, at least) or by comparing it to the missing part of other fossils of the same species or closely-related species (but make sure to scale it up or down in size appropriately!).
In the case of soft tissues, this is where comparisons to close living relatives and the use of Lagerstätten becomes important. For some organisms, marks on the hard tissues allows us to reconstruct soft tissues which are closely associated with them. An example would be the muscles in vertebrates. Muscles attach to bones, so for fossil species we can fill in the missing parts based on the muscle pattern of living relatives. The same can go for structures like eyeballs, brains, etc., which have a close association with hard tissues.
In other cases, though, it is more difficult. And indeed there are many organisms where we honestly have to be skeptical about treating any one reconstruction as better supported than a host of other equally-likely hypotheses.
Analogies: The oldest method of all is the use of analogies: finding living organisms with similar anatomical structures and inferring a similar life habit. For instance, sauropod dinosaurs have elongated necks, as do giraffes. Giraffes feed high in trees as well as on the ground, so some have argued that sauropods did the same.
Analogies can be weak, however, because often the same structure may be used a number of different ways by the same living group, or by different ways in different living groups. Therefore, they should be treated cautiously.
That said, analogies can be useful. For instance, in one formation from the Early Cretaceous of China there are a set of pterosaur (flying reptile) fossils that are very similar, except that some have tall crests and others have broad pelvic regions (hips) (and in at least one of the latter type, there is an egg present in its body). Thus, a very reasonable explanation is a sexually-dimorphic species rather than two closely-related species.
Trace Fossils: Body fossils are great, but trace fossils are produced by organisms while they were still alive, and give insights into behavior that body fossils do not. For instant, trackways can be used to calculate the speed of locomotion of extinct animals, or whether they travelled in herds or not. Bite marks and other feeding traces, gut contents, and coprolites allow paleontologists to determine which organisms ate which kind of food.
Phylogenetic Inference: As we saw before, we can use the extant phylogenetic bracket to reconstruct soft tissue structures in fossil organisms. We can do the same with behavior. Behaviors shared between the extant bracket (or between an extant relative and securely indicated by an extinct bracket) allows for a Type I inference; behaviors present in just one of the extant bracket may be present in the extinct form, and are only Type II inferences; and those which are not present in either of the extant bracket are Type III inferences and can only be supported by evidence independent of phylogeny.
Biomechanics: Organisms live in the world of Physics, and thus we can use the rules of mechanics to judge the physical abilities of a fossil organism. We know the mechanical strength and properties of tissues such as bone, tendon, muscle, wood, etc. Given the known hard part size, and the inferred soft tissue structures, we can calculate whether a given type of motion or force or the strength thereof would allow for any particular behavior. (Of course, organisms are not REQUIRED to live at the extremes of their biomechanical ability, but this technique at least allows us to see which range of behaviors might be allowable for a particular species.)
1. "Eoanthropus dawsoni": Better known as "Piltdown Man". In the late 19th and early 20th Centuries, early humans had been found in France, Spain, Germany, the Netherlands, and other continental European countries, but not yet the United Kingdom. Furthermore, there was a particular theory of human origins developed at this time (this was before the fossils of earlier hominins from Africa had been discovered): it was thought that humans developed our characteristic powerful brain first, and only later developed a fully upright stance, grasping hands, reduced lower jaw, and so forth.
In 1912 amateur archaeologist and antiquities collector Charles Dawson reported discovering the remains of a human (or human-like) upper skull, as well as much more ape-like teeth and lower jaw, from sediments dated to the Ice Age near the town of Piltdown. These were examined by paleontologist Andrew Smith Woodward, who considered them as representing a new species "Eoanthropus dawsoni" (Dawson's dawn man).This was Britain's major contribution to paleoanthropology, and it conformed to the "big brains first" model, since its braincase was practically modern but its jaw ape like.
Several other paleontologists and paleoanthropologists immediately challenged the idea that these bones and teeth were from the same species, or that they were in fact from the Ice Age. However, many accepted these specimens as genuine because they fit into their views. (This despite the fact that Dawson was linked to hoaxes concerning other antiquities, and there were hints from his specimens supposedly found at Piltdown (like the bone of a wooly mammoth carved to resemble that most British of tools, a cricket bat!)).
However, in the following decades discoveries from Asia and (especially) Africa demonstrate that the "big brains first" model did not fit the vast majority of members of the human ancestral lineage: instead, they tended to be upright first, and only developed big brains later. Suspicion that "Piltdown Man" was a hoax grew; "Eoanthropus" was becoming less and less consilient with the growing body of evidence of other members of the human lineage whose authenticity was not in question.
It was confirmed in the 1950s as new chemical age-dating techniques became available, and showed that these were not a single ancient fossil, but instead a medieval human skull, a more recent orangutan jaw, and fossil chimpanzee teeth, all treated with chemicals to appear fossilized. To this day we do not know for certain who the hoaxer was, although Dawson is the primary suspect, nor the actual motive. But it is clear that many people fell for this forgery because it fit comfortably with their preconceived notions and their national prejudices.
2. "Hesperopithecus haroldcookii": Described in 1922 from a single fossil tooth from Nebraska by paleontologist Henry Fairfield Osborn, he considered this worn molar as the first (and only) evidence to date of a North American species of anthropoid (the group of primates containing the various groups of monkeys, apes, "apemen", and humans). Prior to this, there were no known native North American anthropoids. Unfortunately, a popular newspaper account noted that nothing could be reconstructed form this meagre material, but decided to illustrate it based on Homo erectus! Thus, someone just glancing at the paper would think that an "apeman" had been discovered in the U.S.; indeed, creationist publications refer to this as "Nebraska Man".
However, in 1925 additional work at the site found more specimens, and demonstrate this tooth was not from a primate at all, but from a peccary (NOT a pig, as popular accounts say), and Osborn's colleague William K. Gregory published a paper in Science (America's premier science journal) clarifying this. It was a very reasonable mistake to make: apes and peccaries (and pigs, for that matter) have similar molar shape because they are omnivores. Other scientists recognize that Osborn was making do with the best interpretation with limited material, and he was convinced by Gregory's argument. It's okay to make mistakes in Science, as long as you make them for the right reason.
Here is a series of VERY short videos by TechNYou, called "This Thing Called Science":
"Part 1: Call me skeptical" (2:02):
"Part 2: Testing, testing 1-2-3" (2:30):
"Part 3: Blinded by Science" (2:45):
"Part 4: Confidently Uncertain" (3:01):
"Part 5: Do the right thing" (2:38):
"Part 6: Citizen Science" (3:34):