Thinking of deep time

Marie-Pierre Aubry
Department of Earth and Planetary Sciences, Rutgers University, Piscataway, NJ
email: aubry@rci.rutgers.edu

ABSTRACT: Geological time is one of the most difficult concepts to comprehend, and its measurement is expressed both in relative and
numerical terms. The modern geological time scale constitutes a framework for historical geology, and provides numerical ages in two
forms. One is date; the other is duration. Most geoscientists use distinctive abbreviations (commonly ‘Ma’ and ‘Myr’) to specify date and
duration, but some geochemists do not, using ‘Ma’ (mega-annus) interchangeably and are pressuring journals to follow suit. In an attempt
to determine the usefulness of distinct symbols, the concept of geological time and its different expressions is reviewed, and the method-
ology used to determine numerical ages of crucial datum levels in chronostratigraphy is discussed. Despite the important place given to
dates, geological time scales are foremost about duration of intervals between dates, from which the age of key datum points are deduced.
Durations are quantities, and it may someday be suitable to express them in terms of a non-SI unit. Dates, on the other hand, are merely
geological instants, not quantities, and are not concerned with the SI.

INTRODUCTION to recommendations for a logical and consistent terminology for

describing geological time (Aubry et al., this volume).
Earth Science is concerned with geological time, the double
facetted concept that renders geology a sister-science of astro- MEASURING TIME
physics. The two disciplines overlap in their interest in the solar
system while astrophysics alone addresses the ultimate question Of the four dimensions with which Earth scientists frame Earth
of cosmic time. One facet of geological time deals with the un- history, time is the most difficult to comprehend. We lack intu-
folding of Earth (and planetary) history, in its multiple aspects, ition with regard to the immensity of geological time—how is it
including the chronicle of continents, oceans, climates, and, of possible to actually imagine what 1million years represent, let
prime significance to us, life. The other facet is the recovery of alone 4 billion? As Gould wrote (1987, p. 3), “An abstract, in-
time itself, the elusive concept that is only available to us tellectual understanding of deep time comes easily enough—I
through the material evidence of its passage, in the outer shell know how many zeroes to place after the 10 when I mean bil-
of our planet. While the expression “deep time” McPhee (1981) lions. Getting it into the gut is quite another matter. Deep time is
is appropriate for the concept generally thought to have arisen so alien that we can only comprehend it as a metaphor”. The
during the Scottish enlightenment (Hutton 1795), but which has mountain building cycle, as strata are gradually deposited and
roots as far back as the 11th century with such philosophers as then raised into massive peaks, only to slowly wear away over
the Islamic scholar Ibn Sina, also known as Avicenna (Al Rawi the eons until they are below the sea again—as told in Hutton’s
2002) and his Chinese contemporary Sen Kuo (Nathan 1995). It unconformity at Siccar Point—is possibly the only means to
was, however, in Western Europe that the systematic study of grasp a time span of such magnitude. In itself, however, there is
rock successions as markers of time took hold, starting with the nothing in an unconformity, or any other feature of the rocks,
seminal works of Lyell (1830-33) and d’Orbigny (1850). In the that gives us, a-priori an intuitive sense of duration. It may
brief time since geologists began to think of the age of the Earth therefore be useful to review briefly our human experience of
as reckoned not in centuries but in tens to hundreds of million time as a primer to characterizing geological time.
years—Lyell (1830), for example, reasoned from molluscan ev-
Our everyday experience is informative as to the nature of time,
idence that 240 million years had elapsed since the Ordovi-
which comprises two components: dates and durations. It also
cian—considerable progress has been made in extracting ages
shows that time can be measured in two fundamentally different
and durations from the stratigraphic record. Today the Earth is
ways. A relative method consists in discriminating successive
comfortably dated at ~4.54 billion years (Dalrymple 2001,
intervals of time according to natural or assigned components
2004) and uncertainty in chronologic resolution for most of
that occur in sequential and recognizable order (e.g., Spring-
geological history has been reduced to a few thousand years.
Summer-Autumn-Winter; Dark Ages-Middle Ages-Renais-
sance; Classicism-Impressionism-Fauvism-Cubism), or in pre-
Progress in measuring geological time has been incremental, or post-relation to a unique event (e.g., the Fall of Rome; or a
with advances at different times in different geological disci- given war).
plines, resulting in parallel developments of overlapping, often
irreconcilable concepts, and leading to miscommunication and A quantitative, “absolute”, measurement of intervals of time is
misunderstood terminology. One area of significant misunder- also required to synchronize the complex web of interactions in
standing concerns the terminology used to denote various ex- developed societies, as well as to accurately measure time de-
pressions of geological time, as exemplified by editorial pendent activities. Various methods (standing stones, obelisks,
inconsistencies in major research journals. This paper attempts sundials) were contrived in early civilizations to count the
to reconcile some of these contentious concepts, as background phases of astronomical cycles regardless of duration in order to

stratigraphy, vol. 6, no. 2, pp. 93-99, 2009 93

Marie-Pierre Aubry: Thinking of deep time

recognize specific moments (e.g., solstice, lunar months, ap- Different relative measures of time are available. Several con-
proximate rhythms of the day). The first semi-quantitative cern specialty fields, for instance biostratigraphy, leading to
methods were probably developed in ancient Egypt (Curry, biochronology (from biozones to biochrons), magnetostrati-
1990; Richard, 1998; Lippincott et al., 1999) where the clepsy- graphy, leading to magnetochronology (from magnetozones to
dra, or water clock, divided the day into 24 hours of equal dura- magnetochrons), isotope stratigraphy (with isotopic stages and
tion. The invention of the pendulum-regulated escapement by ages). Appropriate terminology in these different fields is dis-
Huygens (1656) led to the development of reliable, and soon cussed in other papers of this volume.
portable spring-powered clocks and watches which were vital
to the industrial revolution. While the piezoelectric properties Universal relative time in geology is expressed by chrono-
of quartz crystals were first discovered by Jacques and Pierre stratigraphy, a science in itself, founded on clear principles
Curie in 1880, the first quartz clock was not constructed until (Hedberg 1976; Salvador 1984) and more recently regulated
1927 (Marrison and Horton 1928) and the quartz wristwatch, and codified (Remane et al. 1996). Its fundamental concept is
now in almost universal use, first appeared in the 1930s (Astron the spatial (geometric) relationships between rock units follow-
1969; http://www.ieee.org/web/aboutus/history_center/sei- ing the principle of superposition (see Aubry 2007). Not by
ko.html). The microsecond range was reached with atomic right, but by educated consensus (Hedberg 1976), the stage is
beam magnetic resonance (Rabi et al. 1939) and nanoseconds the basic unit in the superpositional hierarchy. Its definition es-
are now counted by atomic clocks (see review in Sullivan tablishes that of the derived series, which in turn defines the
2001). limit of system, and so on to erathem and eonothem. Stages are
the only units defined in the rocks by bounding horizons, in
which physical boundary points in stratotype sections, origi-
If passing time is difficult to measure accurately without so- nally called “golden spikes”, have morphed into the Global
phisticated instruments, then it might be expected that passed Standard Stratotype-section and Point or GSSP (Cowie 1986).
time is even more difficult to apprehend. To wit, it took 126 A boundary horizon corresponds to a geological moment—the
years from the enunciation of the fundamental trio of strati- moment when the horizon was deposited. The interval between
graphic principles by Steno (1669) to the discovery of “deep two successive physical boundaries is thus the embodiment of
time” by Hutton, 38 years more to the first division of the sedi- an inferred interval of time, or “age” (see discussion in Aubry,
mentary record of the European Tertiary into temporal units 2007). In a hierarchy that parallels that of the stage to eonothem,
(Lyell, 1833), another 143 years for the international geological the corresponding ages, epochs, periods, eras and eons are de-
community to endorse a guide to stratigraphy (Hedberg 1976). fined, at base, by the corresponding time horizons taken from
It took 139 years to first give numerical dates to Lyell’s early the rocks.
divisions of time (Berggren 1972), another 10 years to construct
the first integrated time scales (Lowrie et al. 1982; Berggren et Disconcertingly, the introduction of regulation (Remane et al.
al. 1985a), and 20 more years to achieve a relatively stable as- 1996) increasingly threatens chronostratigraphic hierarchy by
tronomically derived Neogene time scale (Lourens et al. 2004). ignoring stages in standard definitions (Aubry et al. 1999), in fa-
In sum, more than 200 years elapsed between the first under- vor of an inverted top-down hierarchy based on a-priori defini-
standing of what we are looking at when we see strata, to reli- tions of higher ranked units, mainly as a convenience
ably measure the extent and duration of time involved since (McGowran et al. 2009). The consequences of abandoning the
their formation. Technological developments have been cru- stratotype-based hierarchy (Hilgen et al. 2006; Aubry 2007;
cial, but their application had to follow key conceptual shifts, Gladenkov 2007) can be seen in the recent ratification by the
with the most significant being the recognition that the history of IUGS (International Union of Geological Sciences) of a pro-
the magnetic field (Heirtzler et al., 1968) and astronomical time posal to create a System/Period for the Quaternary with a
(Hays et al., 1976) could be imposed on the stratigraphic record self-defined boundary at 2.6 Ma that is related only to the con-
for independent determination of the age of specific strati- cept of “first glacial climate”—a concept that has moved repeat-
graphic horizons. edly as opinions have changed (Hilgen 2008). In so doing, the
component Pleistocene Series/Epoch was expanded by 44%
Relative time in Earth Sciences with no justification other than conforming to the imposed top
-down hierarchy, and without regard for the stability of the
many disciplines from paleontology and paleoanthropology to
Clepsydras are no longer in use, and sundials and obelisks now paleoceanography and paleoclimatology, in which Pleistocene
occupy a decorative role, but the passage of time is still evalu- is a key concept and Quaternary is hardly used (Van Couvering et
ated in relative terms, even though the means of knowing every al., in press)—an essentially political maneuver likened to a
single moment of it now exist. Although the dates of occurrence “land grab” (see Mascarelli 2009) with a potential for increased
of many historical events are known with precision, it remains rather than resolved conflict.
useful to express time in relative terms (see above) in order to
communicate concepts and knowledge. If all we knew of hu- But should boundary definitions take full precedence in
man history were the numerical date of each event, we would chronostratigraphy? Common sense in everyday life tells us that
have nothing but a list of disconnected facts. The same is true what happens at 12 noon and 12 midnight is not as important as
for Earth history. Even though we now have the means, in the- what happens in the time between. Certainly chronostrati-
ory, to numerically date every stratigraphic horizon, relative graphic boundaries of rank higher than stage boundaries are as-
time conveys information that numbers do not. The Creta- sociated with extraordinary events. Sometimes the cause of the
ceous/Paleogene boundary is tied to a time of ~66 million years event is abrupt, and its evidence is paramount in locating the
ago, but that number in itself does not resonate like the descrip- boundary. In the bolide impact interpretation of the Creta-
tion of what happened then. Likewise, the term Eocene carries a ceous/Paleogene boundary, for example, what happened just at
meaning that is far more meaningful than when we simply cite the moment the Cretaceous ended [the equivalent of midnight in
the interval between ~55.8 and ~33.9 million years ago. our comparison] is obviously the essence of the boundary. But

94
 Stratigraphy, vol. 6, no. 2, 2009

for most other boundaries, the transforming event is usually FAD (first appearance datum) and LAD (last appearance da-
more prolonged, with its beginning to be found in older time tum) of biotic elements (Berggren and Van Couvering 1978).
and its final end at some younger point as well (e.g., Cao et al.
[2008] for the Permian/Triassic boundary; Aubry and Bord The concept of geohistoric ‘datum’ may not be well understood
[2009] for the Eocene/Oligocene boundary), such that the defin- in broader circles, as the following statement suggests: “There
ing stage boundary may represent some element or interval of the are really no such things as ‘datums in geological time’ except
event, that is hardly significant in itself. relative to time of measurement or some arbitrarily defined
benchmark” (P. R. Renne, Paul de Bievre, Maro Bonardi, Igor
The recent emphasis placed on chronostratigraphic boundaries M. Villa; written communication, 29 June 2009; emphasis
of high rank (above stage) may be linked to the recognition that added). The concept was defined for paleontological events,
each is associated with a major change in the Earth system, as and it remains mostly used in this field (see McGowran 2005).
documented in the fossil record (Raup and Sepkoski 1982; see To paraphrase Holland (1984, p. 149) in his description of a
Aubry et al. 2009). The ICS has considered it essential to pro- GSSP, a paleontological datum is simply a biostratigraphic ho-
vide firm numerical age and correlation tools for such major rizon where “time and rock coincide”. An example would be the
chronostratigraphic boundaries per se creating a false impres- lowest occurrence of a taxon corresponding to the taxon’s evo-
sion that boundaries are dated in isolation from stage-stratotypes lutionary appearance (see Aubry 1995). Datums are thus ex-
(which indicate durations), although in reality this is not the case. pressed interchangeably in terms of stratigraphic position, with
regard to distance from a 0-reference point, and in terms of time,
Numerical time in Earth Sciences with regard to the age of the horizon from which the age of the
specified event is deduced.
Much of stratigraphy deals with the integration of various
datasets that, through correlation of stratigraphic successions, As a stratigraphic horizon, a datum also exists independently of
leads to establishing relative temporal frameworks for geologi- whatever date may be applied. Radio-isotopic dating is impor-
cal reconstructions (see for instance McGowran 2005). These tant, however, precisely because it provides ages for datums,
temporal frameworks (biomagneto chemostratigraphic and which in the broader sense can be defined as stratigraphic hori-
alike) are tied to chronostratigraphy. Yet, despite its essential zons that have acquired a specific significance for their geo-
role in Earth history, chronostratigraphy is lacking in one as- historical, paleontological, isotopic, or other character. A
pect: it is mute with regard to numerical time, without which the chronostratigraphic boundary itself is comparable to a datum: a
rates of geological processes cannot be known. point in the rock (no thickness) that represents a point in time
(no duration) (Harland 1978; see discussion in Aubry et al. 1999;
Numerical time is accessible in three principal ways. One is ra- Aubry 2007; emphasis added).
dio-isotopic chronology, based on decay of relatively unstable
isotopes in geological and archeological materials. Isotopic The inherent uncertainty of radioisotopic dates thus has no con-
chemistry measures radiometric quantities that are converted sequence on what stratigraphers refer to as a datum in terms of
into durations, from which, in turn, a (averaged or estimated) relative spatial succession. Numerical ages are regularly im-
geological age is determined. Radio-isotopic ages are often re- proved, but the best dated mineral would be useless unless it was
ferred to as “absolute” ages despite the inherent uncertainty of precisely placed in its specific stratigraphic context. Like a fos-
measurement as well as differences in controls, calibration and sil, a date once extracted from the rock without reference to its
exponential decay (Faure 2004; Dickin 2005). Different iso- location, is lost for science (see Aubry 2007a).
topes in different minerals may yield different ages for the same
stratigraphic horizon, prompting stratigraphers to specify the Ages and ages
nature of the radioisotopic age (e.g., a plateau age) and the
nuclides involved (e.g., a 40Ar/ 39Ar age of X Ma, a 238U/235U In the Earth sciences, as in every day life, the term ‘age’ refers
age of Z Ma). Another methodology is astrochronology, which to duration. Being ‘age 15’ means being in existence for a dura-
interprets Milankovich –derived cyclicity in sedimentary sec- tion of 15 years. Chronostratigraphic age, either sensu stricto
tions. (i.e., equivalent of “stage”) or sensu lato (broadly applied to a
chronostratigraphic unit of any rank) also designates dura-
Astrochronology proceeds with the direct measurement of time tion—for instance being of “Eocene age” implies existing
in the stratigraphic record through the duration of orbital cycles. within a span of time of about 22 million years and a Cretaceous
Most remarkable is the fact that astrochronology now assists age denotes a span of about 80 million years corresponding to
the intercalibration of the 40Ar/39Ar system (Kuiper et al. 2008; this period (cf. Gradstein et al. 2004).
Hilgen and Kuiper 2009). A third methodology obtains an indi-
Duration is the inherent property of that which endures, even if
rect, if constrained, age by interpolation between horizons of
we are unable to measure the duration beyond a certain point.
known ages (e.g., Aubry 1995; Aubry and Van Couvering
Each chronostratigraphic entity has its own duration, estab-
2004). This is not a measured age, but an indirect/estimated age
lished not by decree or by observation but by tying its beginning
derived from the former. The advantage of the first two method-
and its end to a specific moment, defined by a single physical
ologies lies in their independence from sedimentation rates;
point in the rock record, while realizing that the numerical ages
whereas such rates are needed to adjust the values of indirect
given to the bounding points can only be less accurate than the
ages.
true limits of the age in question. The age of the Creta-
Datums in geological time
ceous/Paleogene (C/P) boundary, for example, is approximately
66 Ma (Kuiper et al. 2008), which means that the stratum mark-
The objective of numerical methods is to determine the age of ing the boundary datum was deposited, as far as we know, about
specific horizons. These may be chronostratigraphic boundaries, 66 million years ago. The precise year may never be known, but
or the stratigraphic expression of widely correlatable features it has been proposed that the bolide held responsible for the end
such as magnetic polarity reversals, isotopic signatures, and the Cretaceous mass extinction may have impacted the Earth in

95
Marie-Pierre Aubry: Thinking of deep time

early June (Wolfe 1991) of that indeterminate year. The LAD from the amount of the subject radio-isotope in the sample and
of planktonic foraminiferal genus Hantkenina, the criterion for the half life of the radio-isotope. The date of the distal Point [c]
the Eocene/Oligocene boundary, is presently characterized in is resolved by algebra with the formula (b-a=c). As [a] = 0,
the GSSP section of Massignano (Italy) at ~33.9 Ma (Hilgen [c]=[b]. Thus in radio-isotopic dating the same value in years
and Kuiper 2009) meaning that these extinctions occurred describes both, duration, a quantity, and the distal point, which
~33.9 million years ago. The Brunhes/Matuyama polarity re- is not a quantity. As noted previously, most Earth scientists
versal boundary is placed at 0.778 Ma (Lourens et al. 2004), in- would use ‘Ma’ for the distal point and ‘Myr’ with the calcu-
dicating that the magnetic field reversed 0.778 million years ago. lated value, to distinguish the two components of time.
Each age in these different cases expresses the timing of an event,
whether sedimentologic, cosmic, evolutionary or magnetic. There can be only one distal point that is 66 Ma with reference
to a proximal point set at 0 Ma. On another hand, a literally infi-
Chronostratigraphic ages and numerical ages thus differ in a nite number of other durations of 66 million years are included
fundamental way. One refers to a duration, the other to a dis- in the Earth’s 4.6 billion years because there is an infinity of pos-
crete stratigraphic horizon. They also differ in their stability. sible reference points. This is generic duration, in contrast to
Once a chronostratigraphic unit has been defined by physically specified (or ear-marked) duration. Radioactive decay is unidi-
fixed boundaries, its true duration remains unchanged. In con- rectional (i.e., non-repetitive) albeit non-linear, explaining why
trast, numerical ages may vary considerably, even in measure- radio-isotopic dates and durations, both being measured from
ments on the same material, let alone in different samples the present, are seen as interchangeable.
measured in different laboratories with different tools (see
above). For this reason numerical ages are often explicitly char- Other durations are based on mathematically predictable cycles,
acterized by method, whether radio-isotopic, astronomical, or and thus have constant uncertainty values, rather than percentage
estimated. As the numerical ages of chronostratigraphic datum error that increases in absolute value with the increase of mea-
points are progressively adjusted and with astrochronology be- sured time. Astrochronology, which depends on well-docu-
ing extended to older and older time (Hinnov and Ogg 2007), the mented long-term orbital interactions, is the most advanced and
estimated duration of chronostratigraphic units changes as well, exact of such cycle-based time scales. Where the sedimentary
asymptotically approaching the actual value. record can be firmly tuned to the computed astronomical solu-
tions, the evidence of orbital cycles can be numbered in a linear
Numerical ages of stratigraphic horizons (and by inference sequence starting from cycle 0, in the present day (Lourens
events; see Aubry 2007) are the equivalent of dates in a calen- 2004; Pälike et al. 2006). In other parts of the record, however,
dar because they refer to a time counted in years from a starting long series of orbital cycles have been identified in sections that
point. 1 Ma, as a date, is constructed in the same way as Year can only be provisionally related to the anchored sequence, pro-
AD 1 (Anno Domini), or year AD 1669, or 15 June 2009. In ver- viding temporary but still useful local frameworks (Olsen and
nacular language‘year’ is often omitted in the writing of dates, Kent 1999; Cramer et al. 2003; Westerbroek et al. 2007). These
which can be simply AD 2009 when the context is clear, but floating time scales are methodological proof that durations and
‘Ma’ is, so far, customary for all geological dates. points are two independent variables. As they await to be an-
chored by satisfactory astronomical solutions, floating time
Ma means Mega-annus (from Greek: megas large; and Latin:
scales are also indication that the future of the geological time
annus year, Berggren and Van Couvering 1979) – not Mega-an-
scale lies at least equally well with the powerful resolution of
num, and also not Million years ago (see Aubry et al. this vol-
orbital periodicities as with the improvements in radio-isotopic
ume). The symbol ‘Ma’ is used expressly in stratigraphy to
dating.
differentiate ‘date’ from ‘duration’ which is commonly noted as
‘Myr’, ‘my’ or ‘m.y.’, based on abbreviations for ‘million years'. Even in radio-isotopic dating it is possible to demonstrate that
As a formal term, the symbol Myr is preferable, since in SI duration and points are different. Using a strict logic, the zero
symbology lower case ‘m’ stands for ‘milli’ or thousandth. The point of a historical progression should be at its beginning. Plac-
importance of the distinction between date and duration, rather ing Year 0 of geological time at the beginning of the solar sys-
than to use a symbol such as ‘Ma’ for both is important to clarify. tem, a duration of 4.54 billion years (4.54x103 million years)
Points and duration means that today would be 4.54 Ga (giga-annus; 4.54x 103 Ma).
The oldest rocks on the planet (O’Neil et al. 2008) would be 26
To restate the obvious, duration is an interval of time between Ma, not 4.28 Ga. A duration of 66 Myr, measured from today,
two moments, i.e., two points in time. It follows that any con- would date the C/P boundary at 4.474 Ga. This makes clear one
sideration of time involves three parameters, a proximal point, (and perhaps the least problematic) of the difficulties associated
an interval, and a distal point. The greatest duration for Earth sci- with the logical measure of time from the birth of the solar sys-
ences is 4.54 billion years, from the time of the formation of the tem: i.e., the inconvenience of the very large numbers for the
solar system to the time of today. Intermediate points in this Phanerozoic Eon, which would begin at 3.998 Ga. Neverthe-
4.54 billion years temporal continuum are necessary to compre- less, there are some interesting illogical consequences when
hensively describe Earth history. These points are the counter- Earth scientists instead measure time back from the present, but
parts of the dates in day, month and year of calendars upon do not assign negative numbers to the past dates. For instance,
which historians rely to recount the human adventure. How are an event that occurred 3 million years after the C/P boundary is
these intermediate points determined? 63 Ma. This is a smaller value than 66 Ma, even though a dura-
tion of 3 Myr has been created. Obviously, our geological calen-
In the time scale some of the intermediate points are the direct dar does not obey a distributive law. But durations do: 3 Myr
products of radio-isotopic dating (see discussions in Berggren before and 3 Myr after the C/P boundary add up to 6 Myr. Com-
et al. 1985b, 1995). In radio-isotopic dating two parameters are parable illogic occurs in our dealing with the passage of time.
known: 1) the Present, or 0 million year, which is the proximal By setting Year 1 of calendars some time in the past—e.g., 2009
Point [a], and 2) the duration [b], which is the age calculated years ago in the year AD when Jesus was born— we measure

96
 Stratigraphy, vol. 6, no. 2, 2009

passing time correctly as an increasing number of years. Yet, a tage of two ways of perceiving time, as we reconstruct Earth
30 year old person is not a person who was born in Year AD 30, history, is felt by every professional, and to discard, in the name
but one born in AD 1979 — one more example of the difference of uniformity, the intellectual tools that have evolved to manage
between date and duration. this information is hardly an act of progress.

‘Myr’ and ‘Ma’ The writing of this essay was prompted by an e-mail exchange
between Nick Christie -Blick and Lucy Edwards concerning the
The distinction between durations that are specified with re- terminology and abbreviations to be used in earth sciences to
spect to beginning (0 Ma) and durations that are not is extremely characterize time units. I thank them for involving me as they
important, and far from being a whim of stratigraphers it is a broadened the discussion among colleagues. I am grateful to
matter of scientific logic and rigor. The year is the unit of time W.A. Berggren, J.A. Van Couvering, D.V. Kent, K.G. Miller,
in both cases, but it is constrained as a specific date in the first D. Owens, C. Swisher, for discussion and/or comments on this
instance, while in the latter it is unconstrained. This calls for the manuscript. I am deeply indebted to J.A. Van Couvering for his
use of two distinct symbols, one for duration, any duration, the careful editing of the manuscript that has helped clarify a con-
other for dates. While the symbol ‘a’ has been widely proposed, tentious issue. Finally, I a m thankful to the Commissioners of
if never formally adopted in SI, to simply mean ‘year’ (cf. the North American Commission on Stratigraphic Nomencla-
Aubry et al., this volume), the use of ‘Ma’ to denote dates (age ture for encouraging me to contribute this paper to this special
of a point in millions of years before present) is so ingrained in issue of Stratigraphy on the North American Stratigraphic Code
the Earth sciences (including in geochemistry) that it would be and related subjects.
wise to retain it for this purpose. The symbol ‘Myr’ and other
multiples of ‘yr’ are available for duration, whether specified or REFERENCES
unspecified, and could be used as the symbol for a non-SI unit if
agreement is ever reached on the meaning of ‘year’. AL-RAWI, M. M., 2002. The contribution of Ibn Sina (Avicenna) to the
development of Earth Sciences. Foundation for Sciences, Technol-
ogy and Civilization, 4039. 12 pp.
CONCLUSIONS (pdf; see http://en.wikipedia.org/wiki/Deep_time).
Our relationship with passing time is a learned experience be-
ginning in childhood. However, as soon as a watch is attached AUBRY, M.-P., 1995. From chronology to stratigraphy: Interpreting the
stratigraphic record. In Berggren, W. A., Kent, D. V., Aubry, M.-P.
to a wrist, the illusion sets in that time has become tamed, and
and Hardenbol, J., eds., Geochronology, time scales and global
the essence of time is no longer a concern. Yet, the passage of stratigraphic correlations: A unified temporal framework for an his-
time helps us understand the nature of past time, and the mean- torical geology, 213-274. Tulsa: Society of Economic Geologists
dering ways that have led to increasingly precise measurements and Mineralogists. Special Volume 54.
of passing time also assist us in understanding the tortuous ways
that made it possible to grasp the immensity—not the eternity ———, 2007. Chronostratigraphic terminology: Building on principles.
of Earth and planetary time. The measurement of past time is Stratigraphy, 4: 117-125.
not a blue ribbon dotted with check marks every so many mil-
lion of years depending on the availability of datable radioac- AUBRY, M.-P. and BORD, D., 2009. Reshuffling the cards in the photic
tive elements in rocks. Its measurement is a web of zone at the Eocene/Oligocene boundary. In Koeberl, C. and
Montanari, A., eds., The Late Eocene Earth — hothouse, icehouse,
interconnected (correlated) stratigraphic horizons and units,
and impacts, 279–301. Boulder, CO: Geological Society of America.
each endowed with a precise significance, contributing to a Special Paper 452.
mixed calendar of relative and numerical time, just as in a cal-
endar of years and holidays. Radio-isotopic dating, no longer AUBRY, M.-P. and VAN COUVERING, J. A., 2004. Buried time:
the unique means of accessing numerical time, has been largely Chronostratigraphy as a research tool. In Koutsoukos, E., ed., Ap-
supplanted by astrochronology for measuring more recent geo- plied stratigraphy. Cambridge: Cambridge University Press, p.
logical time, and increasingly more of the earlier record as well. 23-53.

Contrary to first impression time scales are not made of dates, AUBRY, M.-P., BERGGREN, W. A., VAN COUVERING, J. A. and
but of durations. Durations are identified by delimiting datum STEININGER, F., 1999. Problems in chronostratigraphy: Stages, se-
points so that other durations can be known. Quantification of ries, unit and boundary stratytpes, GSSPs, and tarnished Golden
Earth history by the conversion of datums into numerical dates Spikes. Elsevier Earth Science Reviews, 46: 99-148.
that are referenced to the present put durations into a common AUBRY, M.-P., BERGGREN, W. A., VAN COUVERING, J. A.,
frame of reference. To sum up, the numerical ages on the left MCGOWRAN, B., HILGEN, F., STEININGER, F. and LOURENS,
side of time scales —the ages of GSSPs, biostratigraphic L., 2009. The Neogene and Quaternary: chronostratigraphic compro-
events, magnetic reversals, isotopic peaks, mass extinctions, mise or Non-overlapping Magisteria? Stratigraphy, 6: 1-16.
i.al., are a special class of durations that extend backwards from
the present as a proximal date, to a unique instant of geological BERGGREN, W. A., 1972. A Cenozoic time scale—some implications
time at the end (logically the beginning) of the specified dura- for regional geology and paleobiogeography. Lethaia, 5: 195-215.
tion. Considering the rates of geological processes, these are
measured in periods with specified durations of millions or BERGGREN, W. A. and VAN COUVERING, J. A., 1978. Biochron-
thousand years, and here again these durations cannot be de- ology. In Cohee, G. V., Glaessner, M. F. and Hedberg, H. D., eds.,
Contributions to the geologic time scale: 39-56. Tulsa: American As-
scribed or calculated based on a single point with the value sociation of Petroleum Geologists. Studies in Geology No. 6.
‘Ma’. The distinction, then, is fundamental and openly clear:
that when duration is not confused with age, as in normal ———, 1979. Quaternary. Treatise on invertebrate paleontology, Part
geohistorical contexts where the measurement of a quantity of A. Introduction. A505-A543. Boulder, Colorado: The Geological So-
years can begin at some datum other than the present day, then ciety of America, Inc and Lawrence, Kansas: The University of Kan-
the distinction between date and duration is clear. The advan- sas.

97
Marie-Pierre Aubry: Thinking of deep time

BERGGREN, W. A., KENT, D. V. and FLYNN, J. J., 1985b. Paleogene HILGEN, F. and KUIPER, K. F., 2009. A critical evaluation of the nu-
geochronology and chronostratigraphy. In: Snelling, N. J. Ed., The merical age of the Eocee-Oligocene boundary. In Koeberl, C. and
chronology of the geological record, 141-195. London: The Geolog- Montanari, A., eds., The Late Eocene Earth — hothouse, icehouse,
ical Society. Memoir 10. and impacts, 139-148. Boulder: Geological Society of America. Spe-
cial Paper 452.
BERGGREN, W. A., KENT, D. V., FLYNN, J. J. and VAN
COUVERING, J. A., 1985a. Cenozoic geochronology. Geological HILGEN, F., AUBRY, M.-P., BERGGREN, W., VAN COUVERING,
Society of America Bulletin, 96: 1407-1418. J. A., MCGOWRAN, B. and STEININGER, F., 2008. The case for
the original Neogene. Newsletters on Stratigraphy, 43: 23-32.
BERGGREN, W. A., KENT, D. V., SWISHER, C. C., III, and AUBRY,
M.-P., 1995. A revised Cenozoic Geochronology and chronostrati- HILGEN, F., BRINKHIUS, H. and ZACHARIASSE, W.-J., 2006, Unit
graphy. In Berggren, W. A., Kent, D. V., Aubry, M.-P. and stratotypes for global stages: The Neogene perspective. Earth Sci-
Hardenbol, J., eds., Geochronology, time scales and global strati- ence Reviews, 74: 113-125.
graphic correlations: A unified temporal framework for an historical
geology, 129-212. Tulsa: Society of Economic Geologists and Min- HOLLAND, C.H., 1984. Steps to a standard Silurian. Proceedings, 27th
eralogists. Special Volume 54. International Geological Congress, Moscow, 127–156. Utrecht:
VNU Science Press.
CAO C., WANG W., LIU L., SHEN S. and SUMMONS, R.E., 2008.
Two phases of d13C anomalies in the latest Permian: Evidence from HUYGENS, C., 1656. Die pendeluhr, Horologium oscillatorium. hrsg.
the terrestrial sequences in northern Xinjiang, China. Earth and von A. Heckscher und A. v. Oettingen, mit einem bildnis und 113
Planetary Science Letters, 270: 251-257. figuren im Text [ex. 1913, . Leipzig: W. Engelmann. Series
Ostwald’s klassiker der exakten wissenschaften, nr. 192]. 266 pp.
COWIE, J. W., 1986. Guidelines for boundary stratotypes. Episodes, 9
(2): 78-82. HUTTON, J., 1795. Theory of the Earth, with proofs and illustrations.
Edinburgh: William Creech.
CRAMER, B. S., WRIGHT, J. D., KENT, D. V. and AUBRY, M.-P.,
2003. Orbital forcing of d 13C excursions in the late Paleocene-early KUIPER, K. F., DEINO, A., HILGEN, F. J., KRIJGSMAN, W.,
Eocene (Chrons C24n-C25n). Paleoceanography, 18: 1097, doi: RENNE, P. R. and WIJBRANS, J. R., 2008. Synchronizing rock
10.1029/2003PA000909. clocks of earth history. Science, 320: 500- 504.

CURRIE, L. A., 2004. The remarkable metrological history of radiocar- LIPPINCOTT, K., with ECO, U., GOMBRICH, E. H., and others, 1999.
bon dating [II]. Journal of Research of the National Institute of Stan- The story of time. London: Merrell Holberton Publishers Ltd, 304 pp.
dards and Technology, 109: 185-217.
LOURENS, L. J., HILGEN, F. J., LASKAR, J., SHACKLETON, N. J.
CURRY, J., 1990. General history of Africa II. Berkeley: University of and WILSON, D., 2004. The Neogene Period. In: Gradstein, F. M.,
California Press, and the United Nations Educational, Scientific and Ogg, J. and Smith, A., A geologic time scale 2004, 409-440. Cam-
Cultural Organization (UNESCO), 418 pp. bridge: Cambridge University Press.

DALRYMPLE, G. B., 2001. The age of the Earth. Stanford, CA: Stan- LOWRIE, W., ALVAREZ, W., NAPOLEONE, G., PERCH-NIEL-
ford university Press, 474 pp. SEN, K., PREMOLI SILVA, I. and TOUMARKINE, M., 1982.
Paleogene magnetic stratigraphy in Umbrian pelagic carbonate
———, 2004. Ancient Earth, ancient skies: the age of Earth and its cos- rocks: The Contessa sections, Gubbio. Geological Society of Amer-
mic surroundings. Palo Alto: Stanford University Press, 247 pp. ica Bulletin, 93: 414-432.

DICKIN, A., 2005. Radiogenic isotope geology. Cambridge: Cam- LYELL, C., 1830, 1833. Principles of Geology, being an attempt to ex-
bridge University Press, 492 pp. plain the former changes of the earth’s surface by reference to causes
now in operation, Vol. I, 511 pp. (1830); Vol. III: 398 pp. (1833).
FAURE, G., and MENSING, T. M., 2004. Isotopes: Principles and ap- London: John Murray.
plications. New York: Wiley, 897 pp.
MARRISON, W. A., and HORTON, J. W., 1928. Precision determina-
GLADENKOV, V., 2007. The new Russian stratigraphic code and some tion of frequency. Proceedings of the Institute of Radio Engineers,
problems of stratigraphic classification. Stratigraphy, 4: 169-172. 16: 137-154.

GOULD, S. J., 1987. Time’s arrow, time’s cycle. Cambridge, Ma: Har- MCGOWRAN, B., 2005. Biostratigraphy: microfossils and geological
vard University Press, 222 p. time. Cambridge: Cambridge University Press, 459 pp.

GRADSTEIN, F. M., OGG, J. G. and SMITH, A. G., Editors, 2004. A MCGOWRAN, B., BERGGREN, W. A., HILGEN, F., STEININGER,
geologic time scale 2004. Cambridge: Cambridge University Press, F., AUBRY, M.-P., LOURENS, L. and VAN COUVERING, J. A.,
589 pp. 2009. Neogene and Quateranry coexisting in the geological time
scale: the inclusive compromise. Earth Sciences Reviews.
HAYS. J. D., IMBRIE, J. and SHACKELTON, N. J., 1976. Variations doi:10.1016/j.earscirev.2009.06.006
in the Earth’s orbit: pacemaker of the ice ages. Science, 194,
1121-1131. MCPHEE, J., 1981. Basin and Range. Toronto: Harper Collins Canada
Ltd., 229 pp.
HEDBERG, H.D., Editor, 1976. International Stratigraphic Guide. A
guide to stratigraphic classification, terminology and procedure. NATHAN, S. 1995. Science in ancient China: Researches and reflec-
New York: John Wiley and Sons, 200 pp. tions. Brookfield, Vermont: Ashgate Publishing. Variorum series.
III, 23–24.
HEIRTZLER, J. R., DICKSON, G. O., HERRON, E. M., PITMAN, W.
C. and LE PICHON, X., 1968. Marine magnetic anomalies, geomag- OLSEN, P. E. and KENT, D. V., 1999. Long-period Milankovitch cy-
netic field reversals, and motion of the ocean floor and continents. cles from the Late Triassic and Early Jurassic of eastern North Amer-
Journal of Geophysical Research, 73: 2119-2136. ica and their implications for the calibration of the early Mesozoic

98
 Stratigraphy, vol. 6, no. 2, 2009

time-scale and the long-term behavior of the planets. Philosophical tion. Trondheim, Norway, and Boulder, CO: International Union of
Transactions of the Royal Society of London, series A357: Geologists Scientists, and Geological Society of America. 214 pp.
1761-1786.
STENO, N., 1669 [1916]. De solido intra solidum naturaliter contento
ORBIGNY, A., d’, 1849. Cours élémentaire de Paléontologies et de dissertationis prodromus, Florence. The Prodromus of Nicolaus
géologie stratigraphique. Paris: Masson, vols. 1 and 2. Steno’s dissertation concerning a solid body enclosed by process of
nature within a solid. An English version with an introduction and ex-
PÄLIKE, H., NORRIS, R. D., HERRLE, J. O., WILSON, P. A.,
planatory notes by John Garrett Winter, University of Michigan.
COXALL., H. K., LEAR, C., SHACKLETON, N., TRIPATI, A.
New York: The Macmillan Company.
and WADE, B. S., 2006. The heartbeat of the Oligocene climate sys-
tem. Science, 314: 1894-1898.
SULLIVAN, D. B., 2001. “Time and frequency measurement at NIST:
RABI, I. I., MILLMAN, S., KUSCH, P. and ZACHARIAS, J. R., 1939. The first 100 years”. Washington, DC: Time and Frequency Divi-
The molecular beam resonance method of measuring nuclear mag- sion, National Institute of Standards and Technology.
netic moments. The magnetic moments of 3Li6, 3Li7 and 9F19. Physi- http://tf.nist.gov/timefreq/general/pdf/1485.pdf, p.5
cal Review, 55: 526-535.
VAN COUVERING, J. A., AUBRY, M.-P., BERGGREN, W. A.,
RAUP, D. M. and SEPKOSKI, J. J., 1982. Mass extinctions in the ma- GRADSTEIN, F. M., HILGEN, F., KENT, D. V., LOURENS, L.
rine fossil record. Science, 215: 1501-1503. and McGOWRAN, M., 2009. What, if anything, is Quaternary? Epi-
sodes, in press.
REMANE, J., BASSETT, M. G., COWIE, J. W., GOHRBANDT, K.
H., LANE, H. R., MICHELSEN, O. and NAIWEN, W., 1996. Re-
WOLFE, J., 1991. Palaeobotanical evidence for a June ‘impact winter’ at
vised guidelines for the establishment of global chronostratigraphic
the Cretaceous/Tertiary boundary. Nature, 352: 420-423.
standards by the International Commission on Stratigraphy (ICS).
Episodes, 19: 77-81.
WESTERHOLD, T., RÖHL, U., LASKAR, J., BOWLES, J., RAFFI, I.,
RICHARD, E. C., 1998. Mapping time: The calendar and its history. LOURENS, L. and ZACHOS, J. C., 2007. On the duration of
Oxford: Oxford University Press, 438 pp. magnetochrons C24r and C25n and the timing of early Eocene global
warming events: Implications from the Ocean Drlling Program Leg
SALVADOR, A., Ed., 1994. International stratigraphic guide: A guide 208 Walvis Ridge depth transect. Paleoceanography, 22.
to stratigraphic classification, terminology and procedure, 2nd Edi- PA2201.doi:10.1029/2006PA001322.

99