Over the past few years we’ve talked a lot here about the ascendancy of intelligent design’s scientific research program. I’ve been involved in the ID world now for over 25 years, and this growth has been very exciting to witness. For me personally, going back to my earliest days in ID, it was always a goal to see ID mature into an active and productive research program. This is precisely what has been happening over the past 10 or 15 years. Now, as research director at the Center for Science and Culture, I’ve been blessed with a front-row seat to witness this development — it’s been extremely gratifying and encouraging to experience.
Yet despite the hundreds of peer-reviewed papers that our research program is producing, for some time I’ve felt that an important paper was missing from the literature. A variety of brilliant ID thinkers such as William Dembski, Michael Behe, Stephen Meyer, George Montañez, and others have developed innovative methodologies for detecting design. These are extremely insightful and useful methods — but what has been missing is a “how-to manual” to help fellow scientists in our community appreciate how to use ID to guide research.
Writing this paper has been on my “to-do list” for years. (Literally, it was on the white board in my office.) But over the past couple of years, my colleagues Emily Reeves, Brian Miller, and Jonathan McLatchie and I decided that we would co-write such a paper. Having a fantastic team of scientist colleagues with diverse strengths and areas of expertise was what was needed to bring this paper to life. It got published this past summer in BIO-Complexity, and it’s titled “The Broad Utility of the Theory of Intelligent Design to Guide Scientific Research”.
Boiling ID Down to Its Essence
My longtime vision for the paper was to boil the theory of intelligent design down to the essence of how it works, and then review real-world examples of how scientists are effectively using this method when they use ID to guide their own research. At heart, this method is the scientific method — well known to everyone — consisting of observations, hypotheses, experiments, and conclusions.
So how does ID use the scientific method? It’s a simple method and it goes like this — all taken from our paper — in four simple steps:
Step 1: Use Observations to Recognize the Effects of Intelligent Design. ID research begins with observations of what intelligent agents produce when they act, and what properties are found in objects that are known to have been intelligently designed. These observations reveal the causal powers of intelligent agents.
Step 2: Propose Hypotheses and Expectations About What Should Be Found if Intelligent Design Is Correct. Observations about the causal abilities of intelligent agents can then be directly used to develop criteria which reflect what to expect when an intelligent agent has been at work, effectively allowing us to construct positive, testable expectations about what we should find if intelligent design has occurred. These design-detection criteria effectively represent hypotheses, predictions, or expectations, of what can be anticipated in designed systems.
Before we move on to Step 3, it’s worth pausing to appreciate what you just read. In short, the theory of ID starts by observing what intelligent agents produce when they act. We then use those observations to generate testable expectations or predictions of what should be found if intelligent design is correct. These ID-based expectations or predictions thus look a lot like the observations of intelligent agents that they’re based upon. The table below lists nine specific expectations / predictions that we discuss in the paper:
| Observations about Intelligent Agents | ID-Based Expectations / Predictions of what should be found in nature if ID is correct | |
|---|---|---|
| 1. Operate in a goal-directed manner | → | 1. Goal-directed operation |
| 2. Generate high CSI | → | 2. Presence of high CSI |
| 3. Find rare solutions to complex problems | → | 3. Existence of rare solutions to complex problems |
| 4. Implement a top-down design where multiple parts are coordinated to accomplish a purpose or goal — e.g., hierarchical design where subcomponents are coordinated to contribute to higher functions at each level. | → | 4. Structured with a top-down design where multiple parts are coordinated to accomplish a purpose or goal, including a hierarchical design, wherein subcomponents are coordinated to contribute to higher functions at each level |
| 5. Produce irreducibly complex systems (e.g., machines) | → | 5. Presence of irreducibly complex systems (e.g., machines) |
| 6. Reuse functional components in different systems | → | 6. Reuse of functional components in different systems |
| 7. Optimize competing constraints | → | 7. Performance optimized to meet competing constraints |
| 8. Include components with a function or purpose. | → | 8. Components have a purpose and function |
| 9. Rapidly generate large amounts of new information, including generating new systems which are fully formed and different from previous systems | → | 9. Rapid appearance of new information, where systems are fully formed and designed to be functional from the outset, showing some degree of discontinuity from other pre-existing systems |
Undoubtedly there are other ID-based predictions or expectations that one could develop using this method, but we limited ourselves to nine. The next step, then, is to test whether these ID-based expectations and predictions hold up:
Step 3: Study Nature (e.g., through Experimentation) to Assess if Design-Based Expectations Are Fulfilled. Having developed ID-based expectations, predictions, and retrodictions in steps 1 and 2, investigators can then study features of the natural world whose origin is as-of-yet unknown to determine whether they meet the previously established expectations, hypotheses, and criteria for detecting design. Investigators can also use the criteria for design-detection to yield predictions, expectations, or models that guide research into how natural features operate in the present day.
Step 4: Draw Scientific Conclusions. When experimentation and other studies of nature confirm the observation-based expectations and hypotheses generated from ID, then a conclusion can be drawn that evidence for design has been found, or about how a designed system operates. When other known models are unable or less able to account for the evidence, design can be detected and inferred as the “best explanation.” As is the case with any scientific theory, the conclusions of design are held tentatively, subject to future scientific discoveries.
Using this method turns ID into a positive, predictive scientific theory grounded in observations of the cause of intelligent agency that can be applied to study a variety of different fields.
Three Types of ID research
It’s also important to quote a key section of the paper that explains what we mean by “experimentation,” which also brings out a key point about the scope of ID research:
This step of the scientific method is often called “experimentation,” however in this context “experiment” can potentially mean a wide variety of modes of studying nature, including (but not limited to) classical experiments performed in beakers or using instruments at a laboratory bench, observing animal behavior in the wild, mapping geological layers through fieldwork and collecting and analyzing fossils within those layers, or collecting starlight from telescopes. The theory of intelligent design can thus be applied to experimentally study nature for two general purposes:
(a) As a historical science to study the origins of natural features.
(b) As an experimental science to study the present-day operation of a designed feature.
As elaborated below, these applications of ID theory can be achieved by experimentally testing for features such as CSI, irreducible complexity, function in a system, abrupt appearance, rare solutions to complex problems, coordination of parts, top-down hierarchical design, optimization, and reusage of functional components. When such criteria are fulfilled, and other competing hypotheses cannot account for the data, design is implicated (Table 4). Alternatively, such tell-tale signs of design can be experimentally tested for, and identified, to elucidate how a system works. This scientific methodology allows the theory of intelligent design to be applied in a variety of different fields.
Again, let’s pause to appreciate what you just read: Yes, ID can function as a historical science to study the origins of natural features. Most everyone knows this. But what many people don’t appreciate is that ID can also function as an experimental science to study how natural systems operate in the present-day.
And there’s a third type of ID research — one that arguably focuses on the earlier steps of the scientific method. This is ID theoretical scientific research, which focuses on articulating ID-based predictions (expectations) and developing methods for detecting design (as we saw above).
Where do we go from here? In a further post, I’ll review eight specific examples of fields where ID is using the scientific method to make scientific advancements.