Sizing up the "Biodiversity Crisis": Paleocurves, Measurements, and Problematic Inferences

* Federica Bocchi is a PhD candidate at Boston University. Federica’s research is in philosophy of science, with a special focus on biodiversity science and conservation. Her doctoral dissertation explores the epistemic, methodological, and ethical dimensions of measurement practices in biodiversity science…

Hi Extinct readers! My name is Federica, and like you, I am passionate about old, dead stuff. But my interests are not limited to digging up fossils, even though I admit to having a knack for spotting fossils in the field. As an aspiring philosopher of science, I am fascinated by the creative and challenging enterprise of learning about past ecosystems and life forms, and by the application of this knowledge to inform conservation decisions. (The area of science that uses knowledge of past life to address problems of biodiversity conservation is called “conservation paleobiology.”)

This essay is based on a paper I recently published in the European Journal for Philosophy of Science. The title of the paper is “Biodiversity vs. Paleodiversity Measurement: The Incommensurability Problem.” In the paper, I argue that comparative evaluations of current and past biodiversity are difficult to justify because measurements of current biodiversity and “paleodiversity” are incommensurable. Two things are incommensurable if they have no common measure or otherwise cannot be meaningfully compared. What I suggest is that measures of biodiversity and paleodiversity are incommensurable in this sense. This leads me to conclude that talk of a “biodiversity crisis,” if predicated upon such a comparison alone, needs better arguments. 

Let me spell out these ideas more thoroughly, beginning with paleodiversity.

Paleodiversity and paleocurves

With the exception of Bokulich (2021), philosophers have not paid much attention to paleodiversity. But the concept warrants philosophical scrutiny. Two interesting things about paleodiversity are how it is measured and how it is represented visually. The graphic representation of paleodiversity is usually called a “paleodiversity curve” and plots geologic time on the x-axis against the number of extant species on the y-axis. The result is a curve that displays the ups and downs of life on earth through geologic time, usually beginning with the Cambrian explosion around 540 million years ago. 

The idea of representing the fluctuations of animal and plant diversity in the past goes back at least 170 years. John Phillips (1800–1874), the nephew of the “Father of English Geology” William Smith, is well known among earth scientists for drawing the first paleodiversity curve tracking the increasing diversification of animals from the “Palaeozoic” era (which he named). (See this post by Strange Science for more about Phillips.) As with many great ideas, the time was not yet ripe for this sketch of faunal swings to gain momentum. Perhaps this had something to do with the small controversy initiated by Charles Darwin and his then-provocative ideas about the origin of species. This was happening in the 1860s. 

Phillip’s paleodiversity curve from 1860, showing major reductions in organic life at the intersection of the Palæozoic and Mesozoic, and Mesozoic and Cænozoic, intervals

We need fast forward more than 120 years from the publication of Phillips’ curve to reach the next milestone in the representation of paleodiversity. As a young researcher, J. John (“Jack”) Sepkoski spent years in the library classifying fossil data and analyzing them with then-visionary computer techniques. His hard work resulted in a series of important papers, not the least important of which was his collaboration with David Raup, “Mass Extinctions in the Marine Fossil Record,” published in 1982. In the article, you can find an early version of what has come to be known as the Sepkoski paleodiversity curve: a diachronic representation of the global trends and patterns in marine family richness. The paper is also famous for having identified the so-called “Big Five” mass extinctions (which supplies the basis for current talk of a “sixth mass extinction”). Later, in 1997, Sepkoski produced another paleodiversity curve, changing the taxonomic resolution from families to genera. 

Sepkoski’s paleocurve for genus diversity, with his three great evolutionary faunas (the Cambrian, Paleozoic and Modern) superimposed. Redrawn from Sepkoski (1997)

There are now many more paleodiversity curves available, each displaying fluctuations of marine or terrestrial genera and species during various time intervals (see Alroy et al. 2008; Fan et al. 2020). You can draw a paleodiversity curve for mammals, plants, or any other taxonomic or ecological grouping as long as you have enough fossils and are good at using data analysis techniques. 

A recent paleocurve from Fan et al. (2021). The top graphic shows the trajectories of genus and species diversity between the Cambrian and Triassic; the bottom graphic shows species trajectories broken down by taxonomic group

Learning about paleocurves was one of the “wow” moments of my graduate research. Darwin famously discouraged evolutionary theorists from relying too heavily on fossils to make inferences due to the “extreme imperfection of the geological record” (Darwin 1859, 294). And it is true that fossils are hardly representative of the whole biota of past geological periods. Organisms with shells and skeletons are particularly well-represented in the fossil record, as well as organisms living in certain environments, like the shallow marine shelf. Total rock volume matters too, as do other things (such as luck!). As a result, counting and classifying fossil taxa (obtaining a raw fossil diversity estimate) is insufficient to “open a window” onto past diversity. 

Yet despite all this, paleontologists have successfully developed analytical tools for extracting information from observed fossil assemblages. This has permitted them to estimate deep-time taxic fluctuations more accurately, relieving some of Darwin’s nightmares. Rarefaction techniques, for example, enable paleontologists to compare paleodiversity in samples of different sizes (say one collected in the Cambrian and the other in the Ordovician). But traditional rarefaction techniques underestimate diversity in communities dominated by rare species (such as tropical forests). To solve this problem, John Alroy has developed another statistical technique called shareholder quorum subsampling.  This works by comparing fossil samples based on their completeness (how many actual species have been observed), returning a more “objective judgment about how much more diverse one community is compared to the other” (Chao and Jost 2012, 2544).

So this, in a nutshell, is how paleodiversity is measured. You get the fossils, classify and count them, correct for sampling issues and other biasing factors, and finally, represent them on a paleodiversity curve. Next I will explore how “biodiversity” is measured, and why this makes comparisons with paleodiversity problematic.

“Biodiversity”

Most people accept that a biodiversity crisis is happening, but they rarely reflect on how this crisis is measured. To get a handle on this, let us take as an example another major ongoing crisis: the climate crisis. How do scientists understand and quantify climate change? As the historian of climate science Paul Edwards argues, 

to say that the global climate has changed implies that we know what it used to be. At a minimum, we are comparing the present with some period in the past. We would like to know the details, the trend over time… ideally 100 years or more. And since we are talking about global climate, we need some kind of picture of the whole planet... (Edwards 2010, 4)

In a similar vein, to justify the claim that biodiversity is currently experiencing a crisis, one must compare some measure of the current status of biodiversity to a measure of a past status representing a state of non-crisis.

The most common way of measuring biodiversity involves simply counting species: this is called “species richness.” Yet species richness offers a very limited amount of information to scientists interested in setting conservation priorities. A better measure of biodiversity should include information about how many individuals per species can be found in a sample (“species abundance”), as well as how evenly distributed individuals are among species (“relative abundance”). In addition to this information, phylogenetic data, morphological distinctness, and other features are commonly used to measure the biodiversity of a sample (Maclaurin and Sterelny 2008).*

[* For a philosophical primer on biodiversity, see this article from the Stanford Encyclopedia of Philosophy.]

One of those ubiquitous collages that might lead you to think that biodiversity is just the total number of species on Earth

Biodiversity measures are represented by mathematical formulas called “biodiversity indices.” A biodiversity index summarizes relevant information about biodiversity by attributing a numerical value to a sample so that the “biodiversity score” of a sample can be meaningfully compared to others. For example, the Shannon Diversity index, the most widely used biodiversity index in conservation, represents biodiversity by attributing a numerical value to the richness and abundance of a sample. There has been a proliferation of biodiversity indices in recent decades, each representing specific features of biodiversity, each displaying unique mathematical behaviors, and each fit for particular purposes. 

The thing to notice about all this is that not everyone agrees that biodiversity can be captured by a simple headcount (think back to those paleodiversity curves). Listing species and individuals is seen as a reification of biodiversity, like the case of representing the success of a professional by fixating on how much money they’ve earned in a year, ignoring other information. The idea is therefore to find measures that do not reify biodiversity but instead express its processual character. To give you a taste of an increasingly adopted method to measure biodiversity understood as a process, we can look at the work of ecologist Daniel Faith (1992). Faith has developed an innovative method for measuring biodiversity (sometimes called “Faith’s index”). The first step is to build a phylogenetic tree including the taxa of interest. Then you can determine the “phylogenetic diversity” of a set of species by adding together the lengths of all the branches on the tree that span the members of the set. (This measures diversity because branch lengths are positively correlated with the number of new characters arising along that part of the tree; so phylogenetic diversity quantifies how diverse a set of species is, as opposed to just tallying up species). The take-away, again, is that measuring biodiversity is a complex enterprise, not reducible to inventorying species or counting taxa. 

Incommensurability

Earlier, I characterized paleodiversity as past biodiversity. This suggests that we should be able to compare paleodiversity with current measures of biodiversity to draw quantitative inferences about the biodiversity crisis. However, I argue that there are two things that complicate this comparison, and with it, the inference that the present biodiversity crisis is similar to the crises known from the fossil record (like the “Big Five” mass extinctions). I analyze these two obstacles to the comparison in terms of two different kinds of incommensurability.

The first type of incommensurability I call “data incommensurability.” As paleobiologist Anthony Barnosky and colleagues (2011) have observed, severe data comparison problems arise in the process of assessing how serious the current biodiversity crisis is. For example, due to differential preservation and sampling efforts, the majority of data used to draw paleodiversity curves—both in synoptic studies like Raup and Sepkoski (1982) and Alroy et al. (2008) and in more fined-grained ones like Fan et al. (2020)—comes from marine shelf organisms with specific phenotypic traits like a shell or hard exoskeleton. What this means is that paleodiversity curves most directly represent fluctuations in the richness and abundance of marine taxa; they do not directly represent global swings in past “biodiversity.” Notably, most of what we know about present biodiversity comes from terrestrial species—think of those charismatic megafaunas and lovely birds. Databases of extant marine species are characterized as “data deficient” (e.g., by the IUCN). All this makes direct comparisons between the present and past difficult.

It is also noteworthy that the taxonomic resolution of Sepkoski’s paleocurve (families or genera) differs from the resolution at which biodiversity data are collected today (usually, species). The differential taxonomic resolution also makes the two datasets hard to compare (but see Barnosky et al. (2011) for a possible solution to this problem).

A second type of incommensurability is “conceptual incommensurability.” Consider that to measure a phenomenon, decisions must be made about how it should be conceptualized and operationalized (the latter refers to the measurement operations that are developed to capture the phenomenon of interest). Paleodiversity is conceptualized as taxic fluctuation in time and is represented as a yoyo curve. This is relatively straightforward. However, as we saw in the last section, the conceptualization of biodiversity is far from straightforward, and so is its measurement. When you think about biodiversity, you likely visualize nature’s “endless forms, most beautiful”: the total inventory of life of life on Earth. But as we saw above, biodiversity scientists are increasingly conceptualizing biodiversity not as a count, but as a process. This corresponds to what has historically been a paradigm shift in conservation biology toward protecting ecological processes and diversification patterns instead of a small number of charismatic species. Biodiversity, on this understanding, tracks a quantity we might call evolutionary potential (Maclaurin and Sterelny 2008). This is only an example, but notice that this way of conceptualizing biodiversity implies that you cannot measure it by counting taxa, as paleodiversity studies tend to do.

How do we tell if we are experiencing a biodiversity crisis?

Let’s say you agree with my argument that measures of contemporary biodiversity are distinct from paleodiversity estimates owing to different conceptual frameworks and incompatible data. The next question you will ask is: what should we do about this? Perhaps we need a fresh way of thinking about the biodiversity crisis that does not hinge on the commensurability of biodiversity and paleodiversity. One possibility is suggested by the philosopher Carlos Santana.

Bleached corals in the Great Barrier Reef, which have become a symbol of the “biodiversity crisis”

Among philosophers interested in biodiversity, Santana is known for advancing an eliminativist argument about biodiversity. Basically, Santana thinks that the concept of biodiversity is useless for prediction and explanation in conservation science and, consequently, that we should get rid of it. I don’t think Santana is right about this, but he makes a valuable point that I will build upon here.

Santana’s recommendation is that conservation scientists swap out the biodiversity concept for something more readily measurable, corresponding to what these scientists aim to preserve (for example, taxonomic richness or ecosystem services). Santana calls these “biological values,” which makes explicit the normative considerations embedded in conservation science. Now, take an ecosystem service that can be quantified, such as the value of pollinators in a certain area of the world. To talk about a “crisis” in that bit of nature, you don’t need to compare your value to some deep-past measurement. You can set a threshold based on other considerations and, when that threshold is breached, you are justified in taking action. This could work for other biological values, too, like species richness in a biodiversity hotspot or tree density in the Amazon. Anyway, it provides a principled way of talking about the present crisis that does not hinge on problematic comparisons between the present and the past.

So, to sum up: I have argued that justifying the claim that we are in a biodiversity crisis by comparing measures of the current status of biodiversity to paleodiversity is problematic because the two measurements are incommensurable: they use data that is not easily compared and they conceptualize the focal phenomenon differently. But if we replace talk of “biodiversity” with “biological values,” we can reason about the severity of the present crisis without downplaying its importance, and without relying on inferences based on problematic comparisons.*

[* You might still have questions about another narrative that captures the taxonomic loss are experiencing today, that of the “sixth mass extinction.” We often hear that the world has entered a sixth mass extinction that is comparable to the “Big Five” identified by Raup and Sepkoski. If you are interested in how a mass extinction is defined and how hard it is to comparatively measure today’s rates of species loss, I recommend that you check out my co-authored paper “Are we in a sixth mass extinction? The challenges of answering and the value of asking,” which is forthcoming in The British Journal for Philosophy of Science!]

References

Philosophers on Paleodiversity

Bokulich, A. 2021. Using models to correct data: paleodiversity and the fossil record. Synthese 198:5919–5940.

Bocchi, F. 2022. Biodiversity vs. paleodiversity measurements: the incommensurability problem. European Journal for Philosophy of Science 12, 64. https://doi.org/10.1007/s13194-022-00494-6.

Bocchi, F., Bokulich, A., Castillo Brache, L., Grand-Pierre, G., Watkins, A. 2022. Are we in a sixth mass extinction? the challenges of answering and value of asking. The British Journal for Philosophy of Science. [Preprint: http://philsci-archive.pitt.edu/20953/]

some paleocurve classics

Alroy, J., Aberhan, M., Bottjer, D.J., Foote, M., Fürsich, F.T., Harries, P.J., […] Visaggi, C.C. 2008. Phanerozoic trends in the global diversity of marine invertebrates. Science 321:97–100.

Fan, J.X., Shen, S Z., et al. 2020. A high-resolution summary of Cambrian to Early Triassic marine invertebrate biodiversity. Science 367:272–277.

Raup, D. M., & Sepkoski Jr, J.J. 1982. Mass extinctions in the marine fossil record. Science 215:1501-1503.

Sepkoski, Jr., J.J. 1997. Biodiversity: past, present, and future. Journal of Paleontology 71:533–539.

philosophers (and a scientist) on biodiversity

Faith, D.P. 1992. Conservation evaluation and phylogenetic diversity. Biological conservation 61:1–10.

Maclaurin, J., Sterelny, K. 2008. What is Biodiversity? Chicago: University of Chicago Press.

Santana, C. 2014. Save the planet: eliminate biodiversity. Biology & Philosophy 29:761–780.

Santana, C. 2018. Biodiversity is a chimera, and chimeras aren’t real. Biology & Philosophy 33:1–15.

And here is the SEP article on “Biodiversity,” written by Daniel Faith

other references

Barnosky, A.D., Matzke, N., Tomiya, S., Wogan, G.O.U., Swartz, B., Quental, T.B., … Ferrer, E. A. 2011. Has the Earth’s sixth mass extinction already arrived? Nature 471:51–57

Chao, A., Jost. L. 2012. Coverage-based rarefaction and extrapolation: standardizing samples by completeness rather than size. Ecology 93":2533–2547.

Darwin, C. 1859. On the Origin of Species by Means of Natural Selection, or the Preservation of Favoured Races in the Struggle for Life. London: John Murray.

Edwards, P. N. 2010. A Vast Machine: Computer Models, Climate Data, and the Politics of Global Warming. Cambridge: The MIT Press.

Phillips, J. 1860. Life on Earth: Its Origin and Succession. Cambridge: Macmillan & Co.