Historical Science and Observational Science


February 6, 2014 by earlbellinger

Recently there has been a storm of debate over what difference, if any, exists between historical and observational science. Unfortunately, those championing the idea of there being a difference between the two also have the ulterior motive of wanting to supplant rigorously tested ideas with their brand of dogmatism. As a consequence, the public response has been to reject the notion of there being any kind of divide, which effectively amounts to pouring out the champagne with the cork. Brushing aside fanatical opposition, it is important from the perspective of the philosophy of science to explore these two kinds of science.

In observational science, we have data pertaining to the actual events in question. 1 These data serve as the conditions for exploring various models, which we can then falsify by generating testable predictions about the world.

I show you a coin. I flip it for you many times, recording the sequence of heads and tails. You may then build a model using these observations to declare whether the coin is fair or biased.

In historical science, we do not have data pertaining to the actual events. We only have data pertaining to the aftermath of the events. Therefore we need to discover a set of initial conditions that, when applied to the correct model, give us the present state. But there may be many competing models and initial states, and so it is not necessarily possible to state with certainty which events actually occurred.

A given model and set of conditions that are capable of predicting the present state may be sufficient but not necessary to produce the present state. To assert otherwise would be affirming the consequent.

My lawn is wet. I know from observational science that rain and sprinklers alike are sufficient for producing this behavior. I can observe each of these in action to cause my lawn to become wet. Both conditions are sufficient for my lawn being wet, but only one of them actually occurred. Can you tell me with confidence which actually caused my lawn to be wet?

The models discovered by observational science are essential when doing historical science; indeed, that is the only way to do historical science! The methods employed by Ken Ham et alia that inspired this debate in the first place are, in my opinion and, as has been seen, in the opinion of the mass public alike, very much invalid. However, it is important to acknowledge while the conclusions they drew may not follow from the premises, the premises themselves aren’t necessarily suspect.

Regular science encompasses both observational and historical science, and more often than not, they depend on each other in order to make progress. Moreover, it is not clear whether any discipline ever engages in one exclusively without ever engaging in the other. It is more likely that most activities of normal science lie in a continuum between the two. Importantly, the burden of scientific proof is still required when dealing with both kinds of science — otherwise they wouldn’t be science, now would they?

In summary: in observational science, we have the initial conditions and we seek to find the right model. In historical science, we supply the model and we seek to find the right initial conditions. On the basis of these definitions, I believe they are closely related but different types of scientific activity.

Further reading: Cleland, Carol E. “Historical science, experimental science, and the scientific method.” Geology 29.11 (2001): 987-990.

  1. I’m being somewhat loose with language here; don’t confuse this with an endorsement for direct realism

2 thoughts on “Historical Science and Observational Science

  1. […] nothing I love more than a good skeptical argument. I agree with him that there is a divide between observational science and historical science. His objections are important: when scientists try to infer the past, there is an assumption that […]

  2. Said Simon says:

    I would say that the main problem with your distinction between observational and historical data is that your treatment of it is completely incoherent and cannot possibly warrant any of your conclusions about it. This is because the difference you are trying to elucidate is one of degree rather than of kind, and the difference thus emerges only as a result of methodological standards internal to a particular research tradition.

    Whether you take yourself to be observing the aftermath of causal interactions that have only just taken place or that have taken place millions of years ago is irrelevant to whether or not you have ‘data pertaining to the actual events in question’. In both cases you do. It would be absurd to suggest that, say, stratigraphic data doesn’t speak to ‘actual’ tectonic activity or glaciation patterns or whatever.

    What you seem to be trying to say is that certain kinds of observational practices produce data that are less under-determined than others. The observation ‘my lawn is wet’ could be possible under any number of distinct causal conditions, such as rain, sprinklers, or escaped members of the local psychiatric hospital gaily flinging little teaspoons of water all night. The observation ‘my sprinklers are currently covering my lawn with water’ has a far narrower set of possible conditions.

    Normally when one talks about an historical science, they are talking about taking a certain view of causality that is sensitive to sequence, contingency, or process. Rather coincidentally, I just had to present on it in a social science methodology course ( but for a more general, natural science approach, you might want to start with A Realist Theory of Science. It suggests one way—a wrong way, I think, but wrong in the right direction—of treating data as historical objects in need of interpretation, rather than treating the past as one big dataset.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Earl Bellinger

Earl Bellinger
%d bloggers like this: