This transcript was generated automatically and lightly edited for readability.

Oliver Strimpel: This is Geology Bites with Oliver Strimpel. The geological time scale is organized as a hierarchy of time spans. The longest are the eons, of which we have four: the Phanerozoic, Proterozoic, Archean and Hadean. These are divided into eras; for example, for the Phanerozoic eon we have the Cenozoic, Mesozoic and Paleozoic eras. These in turn are divided into periods, which are what we most commonly use when referring to geological time. The Mesozoic, for example, is divided into three periods, the Cretaceous, Jurassic and Triassic. And finally, the periods are divided into ages. It is the ages that are the subject of this episode, since it is at the boundaries of the ages that we place the golden spikes.

Joeri Witteveen is a philosopher and historian of science with a keen interest in the principles and practice of naming and classification in science. He is an associate professor of history and philosophy of science at the University of Copenhagen. Joeri Witteveen, welcome to Geology Bites.

Joeri Witteveen: Thanks for having me.

Oliver Strimpel: Our subject today is golden spikes. What are golden spikes?

Joeri Witteveen: Golden spikes are specific points in the Earth's rock record that mark the boundaries of intervals of geological time. You just mentioned periods and ages. Strictly speaking, golden spikes don't mark ages; they mark the rock equivalent of ages, which are stages. So formally speaking, they are part of the chronostratigraphic scale, which is the complement to the geochronologic scale; together they form the international geological time scale. You also mentioned that these intervals are organized hierarchically, so that the shorter ones are nested in the longer ones. This is nicely illustrated in the colorful International Chronostratigraphic Chart, which many listeners will be familiar with. If you look at that chart, you will see that next to some of the boundaries of the lowest-level units, the stages, there are symbols: small yellow pins. The presence of a yellow pin next to a boundary on the chart means that somewhere on Earth this boundary is marked with a physical marker, a golden spike. So in essence, these golden spikes are extremely local ways of marking global boundaries.

Oliver Strimpel: I presume we don't actually hammer a spike made of gold into a rock.

Joeri Witteveen: No, that's right; there aren't any golden golden spikes. The spikes are typically made of brass or bronze with some kind of gold-colored finish. And actually, most spikes aren't even spikes. Many golden spikes are just a plaque positioned next to a boundary in an outcrop. And there are some golden spikes which aren't even marked in the landscape. They might only be indicated in the form of an arrow on a photo in the publication that proposes the golden spike location. In those official publications, the proposals for new golden spikes, you won't actually encounter the term "golden spike." There is a formal term, which is Global Boundary Stratotype Section and Point, or GSSP.

Oliver Strimpel: So what exactly is the purpose of a golden spike, then?

Joeri Witteveen: That's a very good question, because if you think about it, it seems rather odd to single out a particular location on Earth to mark a boundary that is supposed to be global in nature and that can also be recognized in many places across the globe. To understand why this practice of designating golden spikes was adopted, it helps to turn to the history of the chronostratigraphic hierarchy, and in particular to the lowest-level unit, the stage. The notion of a stage was introduced back in the 1840s by a Frenchman, Alcide d'Orbigny, who was studying the Jurassic system. He argued that the Jurassic system could be divided up into 10 stages, which, as he wrote, and I quote him, "nature has delineated with bold strokes across the whole earth." D'Orbigny went on to name and describe these stages and their boundaries based on certain fossil assemblages that he had identified mostly in France, but also abroad.

After he published this work, foreign stratigraphers with closer knowledge of the fossil record in their own regions started to complain that these bold strokes that d'Orbigny had written about were absent in their regions, or were not positioned at the same stratigraphic level, or that they actually discerned many more bold strokes than d'Orbigny had identified. So they introduced additional stage names, sometimes distinguishing a lot more stages, and before long the number of stage names and ways of dividing up the Jurassic exploded. By the 1930s, the Jurassic expert William J. Arkell documented the use of nearly 130 different stage names for all kinds of cross-cutting segments of the Jurassic, and he described this as a "stratigraphical Babel."

So some form of standardization was needed to rein in this chaos. This prompted long discussions at meetings of geological societies and led to exchanges in scientific journals, until eventually, and now we're speaking of the 1960s and 1970s, it was agreed that the best way to address the issue was to mark these global boundaries using very local markers. The idea behind it is that the best way to address uncertainty or disagreement about the stratigraphic level of a boundary between stages is to define it by marking a particular level in a particular outcrop as the formal lower boundary for that stage. This meant that the level of a boundary was defined conventionally, that is, by agreement. Which is not to say that there was no empirical component to this approach at all. There certainly is, because the search for a good location to mark a boundary with a golden spike requires a lot of hard work. It can take years, even decades, to complete that effort. And that's also part of the reason why some of the stages on the chronostratigraphic chart still lack such a yellow pin.

Oliver Strimpel: We'll talk a bit later about some of those missing golden spikes. But were people then looking for locations in the depositional record where there was a sharp change, or some clear, visible boundary?

Joeri Witteveen: We might be tempted to think that for boundaries between stages, or higher-level segments of the chronostratigraphic scale, to be meaningful, they should be real discontinuities of geological significance. But actually the guidelines of the International Commission on Stratigraphy, the body that oversees the principles and procedures for assigning golden spikes, state that an obvious boundary should be suspect. These rules specify that golden spikes should preferably be placed in geologically and paleontologically uneventful sections of as nearly as possible uniform sedimentation. And it's not just the guidelines that say that; many well-known 20th-century geologists have said something similar. For example, the British geologist Derek Ager once wrote that the best level at which to place a boundary is, paradoxically, the level at which it's least obvious. And the Canadian geologist Digby J. McLaren said that boundaries should be defined, whenever possible, in an area where nothing happened; that they should be quiet boundaries, geologically speaking.

So in short, they argued that a good site for a golden spike was one with no marked lithological changes, and with an uninterrupted record of several different groups of fossils and other kinds of signals. So it was all about continuity rather than discontinuity. The reason why they argued for this was that this would give those sites the highest possible correlation potential. For example, if a boundary is placed in a section that has a variety of biotic, chemical, magnetic or other signals, then this gives the highest potential for correlating that boundary across the globe, that is, for determining the level of that boundary in other geological locations on different continents, where the lithic and biotic record might be very different.

Oliver Strimpel: Your point about putting it where, so to speak, nothing happens reminds me very much of the earlier episode I did with Bruce Levell, in which he talked about how incredibly difficult it can actually be to detect a gap, even a major gap, in the sedimentary record. And I think what you're saying, therefore, is that one has to take special pains to try and find an area where the depositional record is continuous, so that it can in fact have counterparts in other places around the world, rather than choosing somewhere where a bunch of history is missing, which would then not allow it to be extended elsewhere. When was the system of golden spikes established, and were any alternative ways of defining ages in the geological record considered?

Joeri Witteveen: To go back to the comment by William J. Arkell from the 1930s, who referred to this stratigraphical Babel: that was the kickoff for a decades-long discussion about how to come up with a way of standardizing chronostratigraphy, so that communication would be kept open and we wouldn't end up in this stratigraphical Babel. It took until the 1960s for the golden spike approach to become accepted, and then it still took until 1972 for the first golden spike to be designated, which was for the base of the Devonian. This was actually done in the Czech Republic at a site called Klonk, which is a very appropriate name for hammering a golden spike into the rock record. It was done on the basis of what was then thought to be the site richest in biotic and chemical signals that would help correlate the boundary.

Now, you also asked whether alternatives were considered, and the answer is yes. The alternative was not to fix boundaries by pointing to a particular rock section and saying, this is going to be the boundary, and here I'm going to mark it. The alternative was to say, let's keep the quest for boundaries an open-ended empirical project. These people argued that it would actually be premature to define boundaries conventionally, and that maybe in time, through more research, chronostratigraphers would converge upon genuine natural boundaries, even at the stage level.

Proponents of that alternative said, why not look at how zoologists and botanists are addressing this problem? Because they were also having a problem of a Babylonian confusion of tongues in their domains. But they had come up with a very different kind of solution. Instead of defining the boundaries of biological species, they said, let's just anchor each name of a species: let's attach it to an individual specimen that thereby definitely belongs to that species, but leave the boundaries of our units open for continued discussion, for actual research. And that approach is still followed today. They selected certain specimens as so-called type specimens. A type specimen does not mean that it's a typical specimen, not at all. It's not supposed to be an exemplary member of the species with respect to its traits. All a type specimen does is carry the name of a species around with it. This practice allows taxonomists, these botanists and zoologists, to agree on which name applies to which species, but without necessarily having to agree on what the boundaries of those species are.

And those who disagreed with the golden spike approach said, let's do something similar. Let's not define the boundaries of stages, but let's just find a point somewhere in the middle of each recognized stage, mark that as the name bearer of the stage name, and leave the problem of settling the boundaries as an open-ended empirical project. This, they thought, would allow them to tame this mess. Instead of having, in the case of the Jurassic, 130 competing names, there would be a smaller set of names that everyone could adopt, based on consultation of these type points in chronostratigraphy, you might call them. If you go back to the literature of the 1950s and 1960s, you see that there's actually a lot of discussion about these two approaches and all kinds of variants of them. Indeed, I came across a textbook from the late 1950s in which the authors wrote that it was still very much an open question which "philosophical attitude," as they called it, towards the delimitation of chronostratigraphic units would be the most appropriate. So it certainly wasn't settled at that time. But if you fast forward a decade or two, to the 1970s, it was very much settled by then. The GSSP approach had definitely won out.

Oliver Strimpel: So the boundary paradigm won out, as opposed to putting the spikes in the middle. The ages, or rather the boundaries of the ages, were initially chosen to correspond to physical features in the stratigraphic record that record actual events in geological history, such as a mass extinction, or the first occurrence of a fossil fauna, or climate change. Do we then reposition the golden spikes when we improve upon our estimates of when these events occurred?

Joeri Witteveen: That's a very good question. The answer, I think, is a qualified no. When a golden spike is formally designated and ratified, what follows is a 10-year moratorium on moving it. And even after those 10 years, a golden spike can only be redefined in exceptional circumstances. It often happens that after a golden spike has been formally accepted, geologists come across locations that you could say are objectively better than the site at which the golden spike has been placed, for example because they contain more varied and more precise signals for the boundary that is currently marked by the GSSP. But if the boundary were moved with every such finding, that would risk reintroducing the instability that the GSSP approach was supposed to solve in the first place. So GSSPs are rarely moved.

A recent, quite exceptional case is the redefinition of the golden spike for the base of a stage of the Permian, the so-called Wuchiapingian stage. The golden spike for this boundary was put in South China in 2005, but since 2020 this location has become permanently flooded, so you cannot access the GSSP anymore except with scuba gear. That's been recognized as a good reason to redesignate the GSSP. So less than two years ago, it was moved to a nearby location that is accessible. In the course of redefining this golden spike, the researchers who did so also incorporated new empirical studies, so they moved it to a level that increases the correlation potential of the GSSP. So it's an interesting, philosophically puzzling thing: because of the accident of a site flooding, you move the entire boundary all across the globe, by a few meters in this case.

Oliver Strimpel: The geological time scale is periodically updated, and new versions are issued with different numbers on them, different ages. But I guess what you're saying is that it isn't the golden spikes themselves that are moving; it's the radiometric absolute ages that we assign to them that are being updated. Is that correct?

Joeri Witteveen: Yes, that's correct. That shows the duality of the geologic time scale. As I said, there's the chronostratigraphic component and there's the geochronologic component. If you look at the time scale, the wall chart is usually updated once or twice a year. If you compare the updates, most of what changes is that the numerical ages are fine-tuned. That means that the dating of the golden spikes can also change. Such a date change doesn't mean anything for the golden spike itself. It means that we know more about the numerical age of the level at which the golden spike was placed.

Oliver Strimpel: As you mentioned earlier, some of the ages have not yet received golden spikes to mark their boundaries. One of these is the Cryogenian, which is the penultimate age in the Neoproterozoic, and it was during this age that two episodes of major glaciation that we refer to as Snowball Earth took place. There is an excellent exposure of Cryogenian glacial tillite on the Garvellach Islands off the west coast of Scotland, and some researchers are advocating that the Cryogenian golden spike be placed there. What is the process for establishing a new golden spike, and what motivates people to advocate for one candidate location versus another?

Joeri Witteveen: Usually how it goes is that you have a working group, which sometimes travels around the world to inspect various candidate sites, and they discuss with each other which of those sites is the best one, which has the highest correlation potential. Then they make a proposal that they send to the relevant subcommission, and before a proposal can be accepted, it has to go through several stages of voting. The subcommission has to reach a 60% majority vote in favor. If they reach that, it goes to the International Commission on Stratigraphy, and then it goes to the International Union of Geological Sciences, which can then ratify it.

In that voting, there might be other considerations taken into account than just pure correlation potential. Another dimension is accessibility. A site needs to be accessible to researchers from across the globe. And in some cases politics can even get in the way. For example, the GSSP for the K-Pg boundary, which marks the base of the Danian stage, is at El Kef in Tunisia, in quite a remote place. And there was some discussion about whether that GSSP should be moved to another location, because the political situation in Tunisia was such that it could no longer be accessed. It should also be said that for historical reasons, a lot of the current GSSPs are placed in Western Europe, and there have been discussions about whether that's actually fair. It's not always for good geological reasons that they're in Europe. So shouldn't we have more GSSPs in locations in the Global South? Recently there have been more GSSPs in Asia, in part to address the overrepresentation of GSSPs in Europe. Of course, scientific criteria matter the most in this process, but there can be nice side effects of having a GSSP in a particular location, for example when geoparks are established around them, which can draw tourists and can be used by local authorities to educate visitors about the geological environment.

Oliver Strimpel: In an earlier episode, I talked to Phil Gibbard about whether we should establish an age called the Anthropocene that defines the onset of the global impact of humans on the planet. This is a bit of a special case, in that there is a question as to whether to define such an age at all, aside from the question of where to place a golden spike. So what is the status of the Anthropocene?

Joeri Witteveen: The Anthropocene, geologically speaking, is not happening, for now at least. Last year the Quaternary Subcommission voted down the proposal from the Anthropocene Working Group. One candidate for the golden spike for the Anthropocene was in the middle of the city center of Vienna in Austria. There's a park called Karlsplatz, and one of the proposals was to recognize the golden spike in that location. You can imagine that if that had happened, it would have been very interesting as a tourist attraction in the middle of Vienna. But that's not happening. Also, the candidate that was eventually part of the proposal, Crawford Lake near Toronto, would certainly have developed into a geopark or a geosite. But for now, none of those proposals is going through, because the entire Anthropocene proposal has been voted down.

Oliver Strimpel: Is that a final decision, or is it just pending further discussion?

Joeri Witteveen: It's never final, so it's up to the Anthropocene Working Group to come up with a better proposal. But the reasons that have been given for rejecting it make it unlikely for any such proposal to go through. The reasons weren't just that the golden spike proposal wasn't good enough. The reason for the down vote is that the entire motivation for recognizing a new epoch of the Anthropocene starting around the 1950s wasn't sufficiently well motivated. Which is not to say that the Anthropocene is dead, geologically speaking. You mentioned Phil Gibbard. He's one of the proponents of recognizing the Anthropocene geologically, not as a formal epoch, but rather as an event, just like the Great Oxidation Event. We can have a more informal way of recognizing the Anthropocene.

Oliver Strimpel: In closing, let me point listeners to the episode web page, where there are pictures of actual golden spikes, including the flooded one you mentioned in China that was moved, some golden spike inauguration ceremonies, and golden spike geoparks. Joeri Witteveen, thank you very much.

Joeri Witteveen: Thank you.

Oliver Strimpel: To see pictures and illustrations that support this podcast, go to geologybites.com, where you'll also find a subject matter index of all the episodes. There you can also give me feedback, which I welcome, as well as sign up to get my emails about new episodes.