Grounding Knowledge and Normative

Valuation in Agent-Based Action

and Scientific Commitment

Catherine Kendig

1 Introduction

If the goal of scientists is the acquisition of knowledge, then knowledge in general

and scientific explanation in particular can surely be understood to be the product of
that pursuit articulated as a unified set of observation statements. However, those
focusing on scientific practice, disagree. They claim that knowledge is always under-
stood with reference to a particular context and in light of the actions of an epistemic
agent. Knowledge-making activities are not the result of universal rules for deriving
explanation from facts but the result of critical intersubjective modes of investigation.
A science-in-practice approach turns our attention to the activities of and communi-
cation between scientists in order to understand and characterize the nature of scien-
tific inquiry. As such, it is part of what has been referred to as the practice turn.1 This
refocusing of science on scientific practices highlights the activities that are revealed
when we look at the processes and doings of science by scientists and scientific com-
munities (e.g. hypothesizing, testing, experimenting, theorizing, measuring) rather
than exclusively on the products of science (e.g. knowledge, equations, devices, theo-
ries). The practice turn in philosophy of science is not an apologetic for an and-
practice-too approach to the metaphysics of science. A focus on practice provides a
route to understanding the nature of the world in ways that have been, until recently,
marginalized by aggressive demarcationist interests within traditional philosophy of
science. This form of aggressive demarcationism held that research seeking to inves-
tigate scientific activity and the work of scientists was not really philosophy but just

1
For research within the philosophy of science in practice and sociology of science in practice, see
for instance the work of Hacking (1992, 1995), Dupré (1993), Chang (2004), Rouse (1996, 2003),
Rheinberger (2005), De Regt et al. (2009), Soler (2012), Soler et al. (2014), and Kendig (2016b, c).
C. Kendig (*)
Department of Philosophy, Michigan State University,
503 South Kedzie Hall, 368 Farm Lane, East Lansing 48824-1032, MI, USA
e-mail: [email protected]

© Springer International Publishing AG, part of Springer Nature 2018 41

H. Riesch et al. (eds.), Philosophies and Sociologies of Bioethics,
https://doi.org/10.1007/978-3-319-92738-1_3

[email protected]
42 C. Kendig

sociology.2 Within philosophy, there was no worse criticism than the suggestion that
what one’s putatively philosophical research was up to was not philosophy but was
instead sociology. For many in philosophy, the criticism amounted to disciplinary
slur, and one to be avoided. This active avoidance by philosophers of science to
engage with sociology of science meant that rather than crossing the divides between
philosophy and sociology, many were instead burning any bridges that remained
between them in an effort to protect the discipline from invasion.
Rather than seeking a purely theoretic approach to knowledge or wholly analytic
approach to explanation, a philosophy of science in practice approach focuses on
the activities required in theory-making, knowledge-making, and explaining. Why
is this important? Focusing on an activity-based analysis allows us to “go beyond
thinking about scientific explanation in terms of logical relations between explanan-
dum and explanans, [and] we can consider how the act of explaining arises and how
it is best performed” (Chang 2011: 208). So called pure theoretic approaches that
omit reference to practice succeed in doing so only by assuming science and knowl-
edge acquisition to be a subjectless state of affairs—activities with no actors, under-
standing with no one who understands, and modelling with no modellers. Chang
suggests the solution to this problem is for us to go against the convention of avoid-
ing the second person familiar “you” in our discourse, explanations, and discussions
(Chang 2011). We should (as philosophers, sociologists, and scientists) recover the
importance of what knowledge is as something you or I understand or explain,
rather than as disembodied subjectless answers to questions (Chang 2011, 2016).
Implementing a science-in-practice approach, my aim is to turn attention to the
work of practitioners reengineering metabolic pathways within chassis organisms
such as E. coli. I ask, what, if anything, doing so can tell us about the relationship
between the metaphysical, epistemological and ethical knowledge-making activi-
ties. As such it constitutes an activity-based analysis of scientific explaining and
normative ethical thinking. If successful, it would suggest that an examination into
the practice of science may also provide answers, (or at least more informed lines of
questioning), for other long-discussed problems in philosophy.
I begin with a brief metaphysics and epistemology of classification. Disciplines
have a system of classification that specifies the kinds of things that are the subject
of study for that discipline, e.g. the periodic table of elements, plate tectonics, the
DSM (Dupré 2006). Synthetic biology is no different. To understand the classifica-
tion system one must focus on the processes by which it is used and made. Our
behaviour is informed by what kinds of things we (presuppose) we are interacting
with as well as the goals and values we rely upon in our investigation. In the first
half of the chapter, I examine the nature of scientific inquiry and how the manipula-

2
Chang has also pointed out the tendency of traditional philosophers of science as well as analytic
philosophy in general to use the “just sociology” claim as criticism of practice based approaches
to philosophy: “In the typical analytic philosopher’s picture, the scientist only enters as a ghostly
being that either believes or doesn’t believe certain descriptive statements, fixing his beliefs fol-
lowing some rules of rational thinking that remove any need for real judgment. All the things that
do not fit easily into this bizarre and impoverished picture are denigrated as pieces of “mere”
psychology or sociology” (Chang 2014: 70).

[email protected]
Grounding Knowledge and Normative Valuation in Agent-Based Action and Scientific… 43

tion and use of different techniques within synthetic biology has metaphysical
implications for the notion of parts and wholes, modularity, biological organization
and function. The second half explores the social aspect of scientific inquiry in an
attempt to reveal how normative valuations and ethical judgements are formed
within and across synthetic biology communities. It concludes with a suggestion for
how the epistemological, metaphysical, and ethical modes of inquiry are connected
in networks of practitioners working together as moral agents—and that kinds of
moral objects, epistemic objects, and metaphysical objects are made kinds through
the activities of practitioners within these networks.

2 Categories of Epistemological Activity in Synthetic Biology

Early conceptual work in synthetic biology identified three different knowledge-

making distinctions that exist within the field (see O’Malley et al. 2008; O’Malley
2009;  Morange 2009a). These were intended to distinguish three overlapping
epistemological categories. The categories demarcate diverse knowledge-making
distinctions that lead to different questions being asked, different methods used, dif-
ferent knowledge acquired through these, and different products or outcomes
(O’Malley et al. 2008). The first of these three categories is whole genome engineer-
ing. In this, biological processes and modules are co-opted and redesigned to solve
technological problems for the production of energy or chemicals required for vari-
ous industries. The most publicized example of this was the synthesis of the first
self-replicating, synthetic bacterial cell. This was achieved by Craig Venter and his
company, Synthetic Genomics, who synthesized an entire bacterial genome and
used it to replace the genome of Mycoplasma mycoides thereby creating JCVI-
syn1.0 (Gibson et al. 2010; Hylton 2012). The second is the engineered construction
of functional parts, processes, pathways, devices, and systems (Brent 2004; Endy
2005). The current attempts to modify metabolic pathways in bacteria, yeast, and
algae to generate biofuels are examples of this (Dellomonaco et al. 2010; Georgianna
and Mayfield 2012; Wang et al. 2013). This involves the design and testing of bio-
logical systems and their component parts. Understanding of these functional sys-
tems is born out in their decomposition, manipulation, and co-option. Understanding
of these parts and networks is based on the structure and syntax of the system.
Describing the syntactic structure of the biological network using engineering,
logic, or mathematical models, researchers can gain knowledge of the behaviour of
the module or pathway in terms of its input and output conditions. The third cate-
gory of investigation is synthetic experimental evolution or protocell creation
(Erwin and Davidson 2009; Morange 2009a). In this endeavour, synthetic biology
seeks to understand the process of evolution, biological organization, and the nature
of modularity. Understanding the process also opens up the potential to optimize the
modules, networks, and systems which direct it.
This initial tripartite classification of synthetic biology provided important
knowledge-making distinctions in the modes of research and knowledge based on

[email protected]
44 C. Kendig

these but it also led to further metaphysical questions. For instance, what kinds are
the parts and processes to which these categories refer? And, what kinds of things
are the parts, modules, and systems used within the discipline of synthetic biology?
Pablo Schyfter3 (2012) suggests what he calls an initial “exploration into things and
kinds” offering a “first look analysis” of kindhood for the products of synthetic biol-
ogy and suggests that they fit imperfectly within both technological kinds and natu-
ral biological kinds. He is critical of synthetic biologists for not considering
kindhood and for using engineering as the model and exemplar on which to base
synthetic biology.
Determining what kinds (i.e technological kinds or natural biological kinds or
something else) exist in synthetic biology may ultimately rest on how the discipline
itself and how it is understood in relation to other biological and engineering disci-
plines. The discipline of synthetic biology is sometimes conceived of by practitio-
ners and detractors as a subset of functional biology and as such is characterized as
an application-based, or technology-based mode of understanding that seeks to
explain how something works (see Schyfter 2012 for problems with this view). It has
also been characterized as evolutionary biology due to its attempt (especially in pro-
tocell creation) to answer why-questions: seeking why (rather than how) biological
pathways, devices, and parts work. This difference in the attribution of goals, prod-
ucts, and techniques depending on what types of questions are being asked make the
categorization of synthetic biology as a hybrid or disunified discipline unsurprising.
Its growing epistemic and methodological toolkit seems likely to continue apace--
the result of sourcing and modifying techniques from biology, chemistry, computer
science, mathematics, and engineering (see Morange 2009b; Keller 2009; O’Malley
et al. 2008 for discussions of the discipline-building of synthetic biology).
But relying on this dichotomy of functional and evolutionary biology, of how-
and why-questions does not seem entirely justified—or at least is not always eluci-
datory—within synthetic biology. Knowledge-seeking questions within synthetic
biology do not focus purely on how-questions directed for the purpose of modifying
function. I suggest elsewhere (Kendig 2016a, b) that such dichotomizations fail to
identify the union of how- and why-questions, their mode of investigation, and cat-
egorization and kind-making (or what I’ve called “kinding”) practices typified by
synthetic biology. The making or constructing of material objects, mechanisms,
processes, or pathways; the theoretical construction of models and algorithms; as
well as the devising of repeatable methods and techniques being made in synthetic
biology are all instances of kinding—where kinding is understood as the epistemo-
logical and ontological activities within the practice of synthetic biology and by
which the categories of that discipline or subdiscipline are configured. The outputs
of these practices—the products of diverse synthetic biological research aims, are
exchangeable and repeatable activities that represent, explain, and further advance

3
Schyfter (2012) considers and evaluates the appropriateness of conceiving the products of syn-
thetic biological research as kinds of technological objects. The discussion here differs from his
insofar as I take a practice-based account of kinds that focuses on kinds of modules (see also
Sprinzak and Elowitz 2005; Keller 2009).

[email protected]
Grounding Knowledge and Normative Valuation in Agent-Based Action and Scientific… 45

our understanding of the relation of parts and wholes, the manipulation of develop-
mental pathways, and the nature of biological functioning and organization. This is
exemplified in the engineered construction of functional parts, processes, pathways,
devices, and systems (Brent 2004; Endy 2005). For example, the current attempts to
modify metabolic pathways in bacteria, yeast, and algae to generate biofuels
(Dellomonaco et al. 2010; Georgianna and Mayfield 2012; Wang et al. 2013) rely on
the modification of metabolic pathways through design and testing of biological
systems and their component parts.
Understanding of these functional systems is born out in their decomposition,
manipulation, and co-option. The type of synthetic biology focused on the engi-
neered construction of functional parts, processes, pathways, devices, and systems
is in the business of producing standardized parts, devices, pathways and modules
with known functions. Standard biological parts with known functions are cata-
logued in a number of registries (e.g. Massachusetts Institute of Technology Registry
of Standard Biological Parts). Insofar as these parts are kinds, this practice is the
making of these parts.
In response to Schyfter (2012), I suggest that the imperfect fit of the parts and
processes of synthetic biology into technological or biological kinds is because that
which is kinded is epistemologically heterogeneous. If synthetic biology provides
knowledge of how systems work, then the explanations of nature it provides (e.g.
about how to re-engineer and manipulate them) suggest that it may be more profit-
ably conceived of as a discipline that is epistemologically, ontologically, and meth-
odologically hybrid. In the next section, I characterize the practice of part-making
in metabolic engineering. I suggest that this practice can be couched in terms of
different kinds of modularizing. It can be understood as a study of kinds of synthetic
biological objects in the making and the nature of those things that make up the
discipline (and subdisciplines) of synthetic biology.

3 The Tangle of Modularity

The activities of practitioners whose work focuses on the engineered construction

of functional parts, processes, pathways, devices, and systems appear to be at least
in part based on an underlying philosophical premise—that the world is organized
in a certain way—that living organisms have an organization that is modular and
that the variation that natural selection acts upon results from the recombining of
modules. The premise of biological modularity is an ontological claim that appears
to come out of this particular form of synthetic biological practice. We understand
that the biological world is modular because we can manipulate different parts of
organisms in ways that would only work if there were discrete parts that were inter-
changeable. This is the foundation of the BioBrick assembly method widely used in
synthetic biology (Knight 2003). It is one of a number of methods that allows prac-
titioners to construct and reconstruct biological pathways and devices using DNA
libraries of standardized parts with known functions.

[email protected]
46 C. Kendig

Because of its pick-and-mix approach to disciplinary tool sourcing, synthetic

biology may be particularly well-suited to unpick the tangled meanings of concepts
that have been long-debated within evolutionary and functional biology (Morange
2009a: 374). Modularity, constraints, and convergence can be stripped down and
analysed and given functional parts-language descriptions through synthetic bio-
logical experimentation and manipulation of genomes of model organisms, such as
that of the synthetic biology workhorse, E. coli. In the following, I explore how
metabolic engineering, as an epistemic tool, may be used to disentangle the multiple
notions of two of these concepts—modularity and fitness.
Modularity has been a concept much discussed (Schlosser and Wagner 2004;
Wagner et al. 2007). Although seen to be a central principle of functional biology in
general and synthetic biology in particular, it continues to cause confusion. Within
the emerging discipline of synthetic biology, this inbuilt ambiguity means that con-
text is usually required for the use of the term to confer knowledge about the system
from one practitioner to another.
“Modularity” is used to identify at least two different kinds of claims within
synthetic biology. I refer to these different forms following the philosophical con-
vention for discriminating broad claims from specific claims of concepts using
uppercase and lowercase letters: Modularity and modularity.4
1. Modularity. A general thesis about the nature of all biological organization as
being made up of relatively independent components, like building blocks. This
can be a metaphysical claim: that the organization of living things is Modular; a
claim that applies with univocality to all biological structures. It can be a meth-
odological claim; that the best way to proceed with research is to look for
Modularity. It can be an epistemic claim, that the world is knowable and we gain
understanding by considering it is Modularly organized.5
2. modularity. A specific claim about the parthood of a particular organism, path-
way, device, or process. This conception focuses on the property of being a part
and the metaphysical relationship of individuation or composition of it as it
relates to a whole. The claim of modularity is one about the nature of parts and
wholes. Included under this kind of claim are various sub-characterizations, for
instance: that parts are relatively autonomous, interchangeable, or independent
of the context of other parts, carry out measurable functions, standardisable, or
are non-decomposable.
These two forms frame different spaces of epistemological and ontological investi-
gation. In doing so, they configure the level of biological organization to which the

4
When referring simultaneously to “Modularity” and “modularity” I will use the admittedly awk-
ward “M/modularity”.
5
Each of these claims can be either conceived of from a realist, antirealist, operationalist, or prag-
matic view as well as either one of monism or pluralism. For instance, one might suggest
Modularity is a pragmatic methodology (and that we can be agnostic about whether the world is or
is not really Modular). Someone may justify this claim that it is Modular insofar as our best knowl-
edge comes from a working hypothesis of Modularity that is a heuristic guiding synthetic biology
research.

[email protected]
Grounding Knowledge and Normative Valuation in Agent-Based Action and Scientific… 47

theses of M/modularity apply, and the referent to which they apply (e.g. processes,
objects, relationships, and properties).

3.1 Assembly Methods Affecting Modularity

Synthetic biology is based on a general thesis of biological Modularity (Knight

2003; Morange 2009a). The parts database of biological modules with well-
characterized functions relies on this working hypothesis and the success of syn-
thetically produced networks and parts in practice seems to bear this out at least in
some limited capacity. Taking a practice-focused approach to understanding
Modularity suggests that the assembly method one uses affects the nature of
Modularity of the parts, devices, and pathways one constructs.
The most commonly used assembly method, BioBrick assembly, is based on the
standardization of basic parts with a prefix and a suffix (Knight 2003). Addition of
these two short DNA sequences to any DNA element results in a BioBrick. Two
BioBricks can be ligated together in either order to construct a composite that is
itself a BioBrick containing a prefix and a suffix (Kendig and Eckdahl 2017). The
central concept that enables BioBrick assembly to work is that type II restriction
enzymes XbaI and SpeI produce compatible sticky ends (Kendig and Eckdahl
2017). BioBrick assembly has been a popular method from its inception. However,
it is not without problems. BioBrick assembly depends on purification of DNA frag-
ments after gel electrophoresis. The pairing of the DNA recognition sequences for
XbaI and SpeI during ligation results in a mixed site that practitioners refer to as a
BioBrick scar (Brent 2004; Endy 2005). The scar is six DNA nucleotides in length
and has a defined sequence. There are applications in which the spacing between
two parts must be smaller than six nucleotides or when the BioBrick scar sequence
confounds the function of a device (Kendig and Eckdahl 2017). Another problem
with BioBrick assembly is that once ligated together, two BioBricks cannot be taken
apart for use in another construct (Knight 2003). Assembled BioBricks are compo-
nents that can no longer be swappable or interchangeable. In this way, the genera-
tion of the scar during BioBrick assembly affects the kinds of parts produced and
their degree of Modularity. Being Modular means something different during
BioBrick assembly because of this known outcome.6

3.2 Modularity and Modularizing

The claim of modularity is a claim specific to the compositional structure that parts
and wholes exemplify in a particular system. As such, modularity is an answer to
questions concerning: What is the nature of parts in that particular device or

6
An extended discussion of modularity based on this example is contained in Kendig and Eckdahl
(2017).

[email protected]
48 C. Kendig

pathway?, How are they organized?, or What is the relationship of the composi-
tional structure of them to the whole in question? It characterizes the property and
nature of modulehood. This second kind of modularity avoids the problems that
befall Modularity regarding different assembly methods. The things that are catego-
rized as modular are numerous and may come about due to differing processes of
assembly or categorization. This kind of modularity defines modules by the action
of modularizing. The assembly method used makes parts into modules insofar as
they are chosen by practitioners to be connected or separated from other parts. What
counts as a module (variously understood) and the criteria for modular kindhood
may not be the same for all modules. That is, modularity may not be a property that
can be univocally expressed for all parts—modularity allows for this.
This kind of modularity is not without problems. To say something is a module
or part has typically meant that it bears some sameness relationship, family resem-
blance, or overlapping shared homeostatic set of properties to another part qua
module. In this way, being a module—insofar as it is understood to be a property—
means that it is a property that is instantiable in one way. This means that all modules
insofar as they are modules are homogeneously so. I think this is a mistake. This
modulehood may be a kind differently instantiated. That is, modulehood for one
thing may not be the same as modulehood for another in a radical and non-
comparative way. Some might suggest that this brings into question the legitimacy
of what it is that we refer to when we claim something is a module if there is no
unifying claim. When faced with this proposition, they may prefer to dispense with
all claims of modularity and become eliminitivists. Alternatively, they may embrace
the heterogeneity of modulehood and allow that modulehood may be radically
heterogeneous across all metaphysically parcelled out stuff. Despite its initial
simple understanding, all modules may not belong to the same sui generis category—
modularity may not be something that is univocally expressed.

4 Evolvability

I move now from mapping out some of the conceptual terrain of modularity within
synthetic biology to that of evolvability. How can evolvability be characterized in
synthetic biology7? I suggest that it may be best understood as the capacity of a
population, organism, device, part, or pathway to change over time—that evolvabil-
ity is the facilitated variation of self-organized systems (cf. Calcott 2014). Conceived
in this way it can serve as an umbrella term under which natural, artificial, and
synthetic change over time can be covered.
Evolvability relies on phenotypes being both plastic and stable. Phenotypes of
organisms are plastic insofar as they are responsive to the continual variation within
their environment. Phenotypes of organisms are stable insofar as they may develop

7
For a recent discussion of evolvability and synthetic engineering that is complementary to the one
presented here, see Calcott (2014).

[email protected]
Grounding Knowledge and Normative Valuation in Agent-Based Action and Scientific… 49

reliably despite changes in the resources available to the organisms. Novel phenotypes
arise through the rearranging or recombining of ancestral phenotypes by the organ-
ism (West-Eberhard 2005: 6543). Organisms are able to coordinate the resources
used in their own development because they are:
richly endowed with a capacity for facilitating variation, a small input of random mutation
would lead to a large output of viable phenotypic variation … Instead of a brittle system,
where every genetic change is either lethal or produces a rare improvement in fitness, we
have a system where many genetic changes are tolerated with small phenotypic conse-
quences and whereas others may have selective advantages, but are also tolerated because
physiological adaptability suppresses lethality (Kirschner and Gerhart 2005: 226).

If facilitated variation is the mechanism of natural selection, then it would appear

that the inputs are modules. Organismal modularity maintains robustness of path-
ways whilst allowing recombination of relatively autonomous entities. Facilitated
variation can be understood as explaining the evolvability of organisms through its
self-organizing processes and its relationships to genetic and extragenetic resources.
In this sense, the evolvability of organisms may be underpinned by the specific
modularity thesis that takes the nature of organisms to be organized modularly.
Organisms may change their features depending on a number of factors. These
changes can be attributable to the self-organizing capacities of organisms in select-
ing which resources, in what order, and in what combination, are used as causes of
their own development. The organism could be understood to be the author of its
own variability. Its generic capacities may constrain deleterious variation or enhance
variability that may be advantageous to it. Using these capacities, organisms may
either buffer or enhance any variations to their genetic or extragenetic resources or
perturbations in their inter- or extra-cellular environmental resources (Kendig 2014a).
An organism has both constraints and de-constraints on the variability of its phe-
notype. Its capacity for variation allows it to maintain itself across a wide range of
conditions. The organism’s potential for different developmental variations lie in its
capacity to self-organize its genetic and extragenetic resources. The organism’s
common stock of generic capacities and resources has been understood by Mary
Jane West-Eberhard (2003: 146) as its “phenotypic repertoire”. The organism’s phe-
notypic repertoire includes a number of highly conserved core processes: inter-
cellular signalling and cell sorting (Goodwin et  al. 1993); the capacity for weak
linkage, exploratory behaviour, spatial patterning, compartmentation and modular-
ity of the body plan (cf. Kirschner and Gerhart 2005); the capacity of tissues to
segment or form hollow tubes (cf. Minelli 2003); and the capacity to learn and
remember (West-Eberhard 2003: Ch. 3, 7, 18). A thumbnail sketch of some of the
ways in which variation is facilitated may be useful.
The capacity for weak linkage allows organisms to use a small number of mecha-
nisms relying mostly on the ability to reconfigure these as and when necessary for
different functions. But because these linkages are weak, they are often retraced and
duplicated by other pathways or circuits to strengthen them. The use of weak link-
ages means that there is higher versatility which the organism can use to alter
these pathways when necessary. They offer a socket and plug model that facilitates
interchangeability of different modules. If these were strong, static and unchanging

[email protected]
50 C. Kendig

linkages, the organism would be arguably less fit to adjust to different environmen-
tal situations (Kirschner and Gerhart 2005: 136–7). In the course of evolution small
changes result in wide variability and novelty. Weak linkages can be formed in
many kinds of interactions, such as those between cells, cell populations, tissues,
organs, organ systems, and behaviours. In this way, they confer a standardized way
of connecting different modules to one another. Because these linkages are weak,
individual organs, cells, tissues, or behaviours may also change independently. The
exploratory behaviour of organisms is their responsiveness to different inputs and
outputs. In building such things as neural networks or circuits, the organism begins
by constructing a large number of alternative pathways. The best of these alternative
pathways are selected and stabilized.
A conserved body plan  – one that is retained over a succession of individual
organisms  – enables independent variation of some features without adversely
affecting others. It does so by compartmentation of the body plan into semi-
autonomous functional and structural parts. This modularity of the organism’s parts
increases its capacity for variation as changes in one subunit do not greatly affect
others and thereby reduces the possibility of lethal variations. The more modular
these subunits become, the greater the possibility of variation and specialization of
these structural and functional units within the organism. The generic capacity to
learn through exploratory behaviour provided certain motivating factors, such as the
absence or presence of certain resources needed in the construction of a particular
phenotypic trait or the performance of certain processes or behaviours (e.g. metabo-
lism, reproduction, locomotion, speech), enables the organism to vary its phenotype
over its lifetime in a range of different contexts. Organisms may learn about their
resources and environment by quorum sensing, by chemical cues or by virtue of
their sensory organs. They may manipulate the objects within their immediate habi-
tat, investigate new resources or interact with new organisms (e.g. prey, potential
mates, carers, symbionts) within this habitat, or search for a new one. Certain types
of activities associated with access to food, protection from weather, increased soci-
ality or reproductivity, or fitness may result in some benefit or detriment to the
organism). Activities that effectively increase or substantially decrease resources
are remembered and repeated or avoided. These generic capacities allow organisms
to vary their own development and the phenotypes they construct depending on how
resources are used (Kendig 2014a).

4.1 How Can Engineered Metabolic Pathways Be Maintained?

Building on the discussion of facilitated variation in the last section, I now use this to
return to discuss it in light of particular examples of metabolic engineering in microbial
organisms to explain how pathways can be maintained in synthetic biology. Metabolic
engineering depends on the discovery and investigation of natural metabolic
pathways and the genetic elements that control them, on using that information to
transform suitable host organisms for the desired orthogonal metabolism, and on

[email protected]
Grounding Knowledge and Normative Valuation in Agent-Based Action and Scientific… 51

optimization of the metabolic output. The most high-profile product of metabolic

engineering resulted in bacteria and yeast cells that produce the anti-malarial drug
artemisinin, reducing its cost of production compared to purification from the sweet
wormwood plant (Martin et al. 2003). Another is the recent attempts to engineer cells
to make more of what they may make already, (e.g., single-celled algae producing
oil, or producing a variation on what they would normally produce). This is exempli-
fied in the current research trend in publically and privately funded projects is to
investigate the potential use of various species of algae for biofuel production (Kendig
2014b). Algae produce lipids (oil) as a byproduct of the process of photosynthesis.
The hope is that once the means of harnessing this store of energy is found, algal
biofuels may provide an inexpensive alternative source of fuel that can be produced
with little more than sunlight, carbon dioxide, and a small amount of water. While
advances in synthetic biology research and the understanding of algal alternatives
increase, the scaling up of these fuels requires significant further research resolving
the problems of system optimization and photosynthetic efficiency as well as solving
ways of producing these synthetic biofuels in quantities suitable for commercial use
(Georgianna and Mayfield 2012). Various species of bacteria and yeast have been
considered as particularly suitable for research into the production of synthetic bio-
fuels (Dellomonaco et  al. 2010). Cyanobacteria are another that initially appears
promising. Cyanobacteria, like Synechocystis sp. PCC 6803, can provide a highly
efficient organic system for producing biofuels as they can convert solar energy and
carbon dioxide into biofuel molecules (Wang et al. 2013). Cyanobacteria are particu-
larly good candidates because they possess naturally occurring biosynthetic path-
ways that produce alkanes (a key component of gasoline, diesel, and jet fuel). At
present, research into the use of cyanobacteria for synthetic biofuel production is still
in the very early stages and well behind that of algae research. However, research
focused on reconfiguring these to create an organism that produces alkanes or alkenes
at a rate that is double that of the wild type has been shown to be possible. Synechocystis
mutants have been constructed that overexpress alkane biosynthetic genes (Kendig
2014b). If their photosynthetic pathways were re-engineered, cyanobacteria may be
able to produce alkanes or alkenes at a highly efficient rate (Wang et al. 2013).
But knowing how to sustain the orthogonal pathway over generations requires
resolution of two problems. I refer to the first of these as the optimization problem.
The optimization problem involves figuring out how synthetic biologists can
enhance orthogonal metabolic output when they cannot know all of the variables
that affect it in a host organism and its environment. I call the second problem the
natural selection problem. How can the reduction in fitness in a population of organ-
isms engineered for an orthogonal metabolic pathway be prevented from causing
evolution away from the desired metabolic output phenotype? The native metabo-
lism of all organisms is responsive to a vast array of internal and external variables
including genotype, metabolite availability, pH, osmotic pressure, and temperature
(Eckdahl et al. 2015).8 When synthetic biologists introduce orthogonal metabolism

8
For an extended discussion of this example see Kendig and Eckdahl (2017) and Eckdahl et al.
(2015).

[email protected]
52 C. Kendig

into a host organism, they can optimize metabolic output by taking into account the
variables they know, but must ignore all of the variables they don’t know and some
that they cannot know. Synthetic biology researchers also struggle with the difficulty
of trying to make sure that the organisms they engineer with the ability to carry out
orthogonal metabolism continue to faithfully replicate that genetic capacity. If the
production of the product or the maintenance of the pathway is too onerous for the
cell, the metabolic pathway will (after a number of generations) be disposed of.
Those organisms without the onerously produced product or taxing pathway will be
selected for in preference to those constructed to function by the synthetic biologist.
To sum up in one sentence: synthetic biology’s biggest obstacle is natural selection.

4.2 Is Semisynthetic Evolvability Still Evolvability?

In the previous section, I suggested that evolvability may be understood as the

capacity of an organism, device, part, or pathway to change over time—that evolv-
ability is the facilitated variation of self-organized systems. Conceived of in this
way, evolvability was able to be used as an umbrella concept under which natural,
artificial, and synthetic change over time can be covered. I now want to question the
legitimacy of this as a category or kind. I suggest that synthetic biological research,
as well as the orthogonal products and processes that it creates, sits in a liminal posi-
tion between engineered technological kinds (cf. Schyfter 2012) and natural (or in
the language of synthetic biology “native”) biological kinds. If this liminality exists,
it presents difficulties with regard to the notion of kindhood if kindhood is under-
stood to be a property of the contents of the world as already there for us. Why? To
explain, I’ll use the butchery metaphor often used to explain the nature of natural
kinds as being understood as picking out the natural categories in the world as carv-
ing nature at its joints. Synthetic biological research and the orthogonal products
and processes that arise from it would suggest a notion of kindhood that does not
rely on the pre-carved up contents of the world but instead a notion that involves the
role of the carvers and—following the discussion of how assembly methods may
affect modularity—the ascription of the joints.
As such, the goal of understanding evolvability is directed to the potential to
modify and optimize the modules, networks and systems which direct it. This takes
evolvability to be a capacity that can be defined, represented, measured, intervened
upon and optimized. We might do better to call this “semisynthetic evolvability” as
it differs from natural or artificial evolvability in that the synthetic biology practitio-
ner actively intervenes in the process of facilitated variation and employs the use of
a suite of constructed parts and employs populations of organisms to solve the bio-
logical problem of optimization.
But the question remains: are semisynthetic evolvability and the umbrella con-
cept of evolvability as facilitated variation of self-organized systems legitimate and
veridical categories of evolvability? And how are we to judge them? According to
Morange, “Selection in nature and evolution directed by biochemists and molecular

[email protected]
Grounding Knowledge and Normative Valuation in Agent-Based Action and Scientific… 53

biologists increasingly differ as technological progress is made in synthetic biol-

ogy” (Morange 2009a: 370). At first blush, this would suggest that semisynthetic
evolvability may not count as evolution at all. Later on, however, Morange provides
an alternative account. He argues against his initial suggestion that synthetically
tweaked evolution does not count as evolution. He goes on to argue that current
understanding of evolution is based on an unwarranted assumption of univocalism
that could be challenged by findings in synthetic biology:
some of the extraordinary scenarios of evolution, designed by synthetic biologists, must be
carefully considered by evolutionary biologists. Maybe the models provided by evolution-
ary biologists are too restrictive, leaving too much place to uniformitarianism, the hypothe-
sis that the mechanisms of evolution have always remained the same (Morange 2009a: 374).

Put stronger, perhaps evolutionary biologists have just been getting it wrong.
Natural evolvability is actually more like semisynthetic evolvability. So research
into the latter may yield understanding of the former. Judged according to Morange’s
second suggestion, the umbrella category of evolvability can also be viewed as
veridical insofar as it captures the heterogeneity of synthetic, artificial, and natural
evolutionary mechanisms as well as their concomitant epistemic categories.

5 The Social Aspect of Scientific Investigation

In the foregoing, I have attempted to show that attending to the practices within
synthetic biology—in particular to the reengineering of metabolic pathways—
reveals the generation of knowledge-making categories and the delineation of mod-
ules within the manipulation of biological parts, processes, and systems. I now turn
to the social aspect of the practitioners’ work. Although the specific focus on meta-
bolic engineering is new, doing so relies heavily on Marjorie Grene’s much earlier
identification of the work of scientists as being not just the subject of epistemology
but also of ethical engagement. Writing in 1966, Grene states that scientists’ “work”
is “an instance of the recognition of responsible persons, a performance of the same
general kind as the recognition of patterns, individuals, or persons” (Grene 1966:
223). Grene refers to the social nature of science as work that takes place in “social
enterprises” (Grene 1985). But Grene is not alone in pointing to the ineliminability
of the social within scientific inquiry. It has been characterized more recently in
terms of “systems of practice” (Chang 2012), referred to in virtue ethics as “agent-
based” interactions (Swanton 2003), been described as “social cognition” (Longino
1990), and in some sense, it has been much earlier cashed out in terms of the con-
cept of “conviviality” (Polanyi 1962). I build on this formindable work, focusing on
the social aspect of scientific investigation and the notion of a system of practice in
order to identify the work of practitioners in a reticulated set of knowledge-making
activities.
Following Grene, I take the work of scientists to rely on protocols as sets of
explicit and implicit rules of action. The action of individual agents within the social

[email protected]
54 C. Kendig

group of scientists is not the work of idealized decision-makers but seems to be

biased on individuals whose beliefs are not entirely self-generated. That is, they do
not form in a vacuum but from within a particular context and environment. As
such, scientists working within social enterprises are agents that act from a particu-
lar embedded standpoint. Grene’s view seems underpinned by an agent-based view
that investigates a perspective where action is always from a standpoint, but that the
standpoint is located in social networks of knowledge producing practices. It is that
place from which and as such a precondition for personal action. In this way Grene
appears to rely on Polanyi’s notion of agency (see Polanyi 1962, and in particular,
see Mullins 2009 for a discussion of Polanyi’s notion of agency). For Polanyi, sci-
ence is only considered possible within a society that acknowledges it by affirming
the sorts of questions, types of reasoning, and value of their pursuit, advancement,
and transmission of knowledge. Polanyi (1962: 203) suggests that scientific knowl-
edge is only possible because the sharing of knowledge is itself valued in society.
That it is valued, makes science possible. He suggests that reliance on this social
endorsement is pervasive—it is prerequisite to all epistemic and ethical knowledge.
Polanyi illustrates the pervasiveness:
I cannot speak of a scientific fact, of a word, of a poem or a boxing champion; of last week’s
murder or the Queen of England; of money or music or the fashion in hats, of what is just
or unjust, trivial, amusing, boring or scandalous, without implying a reference to a consen-
sus by which these matters are acknowledged—or denied to be—what I declare them to be.
I must continually endorse the existing consensus or dissent from it to some degree, and in
either case I express what I believe the consensus ought to be in respect to whatever I speak
of (Polanyi 1962: 209).

More recently, the pervasiveness of the social aspect of scientific enquiry as pre-
requisite for objective knowledge is later explored by Longino (1990). Longino’s
interest is in characterizing the nature of knowledge acquisition, but she cautions,
Because we think the goal of the scientist’s practice is knowledge, it is tempting to follow
tradition and seek solutions in abstract or universal rules. Refocussing on science as prac-
tice makes possible the second shift, which involves regarding scientific method as some-
thing practiced not primarily by individuals but by social groups (Longino 1990: 66).

The social nature of scientific practice that Longino suggests here emphasizes
the intimate connection between the character of inquiry as a social and is prerequi-
site for what she refers to as “social cognition”—the idea that scientific inquiry is
not simply an individual pursuit but an epistemic activity relying on the intersubjec-
tivity of critical dialogue within scientific study. In doing so, she follows Grene
(1966, 1985) in fleshing out the nature of what it means to be a social enterprise and
what I suggest Chang (2012) later identifies as a system of practice. According to
Chang, the search for an agent-free or context-free set of categories is simply one
that is ill-founded in philosophy. This is because knowledge is always something
bound within what he refers to as a “system of practice”:
A system of practice is formed by a coherent set of epistemic activities performed with a
view to achieve certain aims…[A]s with coherence of each activity, it is the overall aims of
a system of practice that define what it means for the system to be coherent. The coherence

[email protected]
Grounding Knowledge and Normative Valuation in Agent-Based Action and Scientific… 55

of a system goes beyond mere consistency between the propositions involved in its activi-
ties: rather, coherence consists in various activities coming together in an effective way
toward the achievement of the aims of the system (Chang 2012: 16).

By focusing on systems of practice, Chang foregrounds the activities and goals

of scientists within the pursuit of knowledge rather than with the product of their
activity—the observations statements and their truth value. It is not the veridicality
of statements on their own that tells us what constitutes success in science. In order
to understand truth or anything like it, it is necessary to investigate how these state-
ments function within the knowledge-making activities of scientists. The assessing
of truth or truthlikeness of any statement is valuable but context dependent (Chang
2012: 17–18). Truth can only be assessed within the constraints of the goals and by
the values set out by practitioners that are for instance, working towards solving a
particular problem. That is, what counts as a solution depends on what is deter-
mined to be the problem.
What appears to connect Grene’s notion of social enterprises, Longino’s social
cognition, and Chang’s systems of practice is a sense of the pervasiveness of the
social nature of scientific investigation—a pervasiveness highlighted by Polanyi’s
notion of conviviality, social endorsement, and dissent. I explore the pervasiveness of
the social by looking once again to Polanyi’s notions in order to characterize an
account of ethics. According to Polanyi, it is not the agreement of the community of
individuals that leads to knowledge but the possibility of disagreement (Polanyi
1962). Being in a community provides the environment that makes disagreement
possible. To disagree is to see oneself as being against something and in the possibil-
ity of communicating that disagreement to another. Applying a Polanyi-inspired
approach to normative ethics, I suggest that normative ethical evaluations are the
result of loops of ethical reflectiveness of agents in a multi-agent network of practi-
tioners who construct the grounds for ethical knowledge through their intersubjective
judgements. That is, normative ethical evaluations rely on what people think, what
they do, how they do it, and how they communicate it to others. These evaluations are
dependent on the epistemic capabilities of individuals multiply instantiated in net-
works of practitioners. In synthetic biology, these networks include human agents,
but also their physical manipulations (e.g. measuring, weighing, running gels), math-
ematical modelling, proxied or remote tool use, objects of study (e.g. chassis organ-
isms, BioBricks), and the extended social communities that they work within (e.g.
research networks that span multiple institutions, iGEM competitions, international
research networks). They are spatially, socially, and temporally extended.

5.1 Extended Agency Ethics

In the remaining, I briefly consider how synthetic biology practice reveals valua-
tional as well as the epistemic and ontological categories of research. I suggest that
ethical categories, like the knowledge-making practices, are formed alongside these

[email protected]
56 C. Kendig

within socially extended research groups and sketch an approach to understanding

the generation of ethical knowledge. I refer to this form of ethical knowledge as
“extended agency ethics” to emphasize the intersubjective nature of this social
approach. Extended agency ethics can be understood as taking a naturalistic agent-
based approach to ethics. I clarify the descriptors “extended” and “agency” one at a
time before bringing them together. I refer to this approach as “extended” because
this view relies on the extended-mind thesis in cognitive science (Clark 1995; Clark
and Chalmers 1998). According to the extended mind thesis, thinking is not exclu-
sively something that happens in the brain, it is something that is spatiotemporally
extended (Clark and Chalmers 1998; Clark 2010). Mind includes brain but also
includes tools used to aid thinking (e.g. language, culture, pencils, calculators,
search engines, mobile phone apps, and other people). Knowledge is not just epis-
temologically embedded according to this view, but it shapes and is reciprocally
shaped by our experiences in the world:
we use intelligence to structure our environment so that we can succeed with less intelli-
gence…it is the human brain plus these chunks of external scaffolding that finally consti-
tutes the smart, rational inference engine that we call mind (Clark 1998: 180).

I use this extended mind thesis to suggest how ethical decision making is inelim-
inably connected to the research practiced in groups—but these are groups of indi-
viduals capable of acting. I refer to “agency” in the description of the approach to
indicate that the role of the individual agent. This agent based view of action implies
that we, as agents, are the locus of our activities, e.g. measuring, mapping, or run-
ning gels, in order to effect changes beyond merely measuring, mapping, or running
gels. Agents act for and towards a purpose beyond those movements. They are able
to do so because they have certain causal knowledge of the result of the performance
of these actions.9 The agents’ activities are not reducible to the events of measuring,
mapping, or running gels because it is the person’s running a gel and not the gel’s—
to assume that these are the same thing is to reduce the agency of the agent to an
event. Keeping in mind the ubiquity of the social as expressed by Polanyi, Grene,
Longino, and Chang, extended agency ethics does not assume that intentionality is
the sole domain of the individual nor that the decisions of the individual are autono-
mous. Individual scientists’ work is generated through a social process and forms a
kind of interactive extended agency. That is, their research activities are extended
over the members of the research group and coordinated in virtue of their shared
intentionality in the form of extended cognition. These shared intentions provide the
grounds for normativity in scientists’ social research interactions. Robert Wilson
has recently suggested a view of extended agency akin to the one I have developed
here and elsewhere, (Kendig 2016a), but within a separate context.10 He provides a

9
I follow Hornsby’s account of irreducible agent causation here, (see Hornsby 2004: 11–14 for
further discussion of agency in philosophy of action and Lowe 2009: 196–201 for further distinc-
tion between agent causation and event causation).
10
Wilson’s (2018) discussion of normativity is given in the context of the eugenics movement and
in particular, within a critical analysis of the cognitive processes that lead to the marking of certain
human variation as deficient and other variation as preferred within scientific practice.

[email protected]
Grounding Knowledge and Normative Valuation in Agent-Based Action and Scientific… 57

particularly helpful way of understanding how sociocognitive intentions can be con-

strued as being the basis for normativity:
Normativity exists when there is a distinction between a correct, proper, or appropriate way
for a process, event, or outcome to turn out, and an incorrect, improper or inappropriate
such way. Like extended cognition, normativity arises in and through both non-human and
human cognition; it is not solely a feature of our own species’ activity. [L]ike extended
cognition, the most familiar and robust forms of normativity are those that are the product
of distinctly human practices and institutions that presuppose a kind of shared intentional-
ity…So we have a kind of externally mediated, cognitively driven normativity, and it con-
stitutes an important feature of human social life (Wilson 2018: 130).

Agency is, therefore, not restricted to the individual or to a limited number of

individuals. It can extend to the social research network one participates within with
its requisite values and practices. Extended agency provides a conception of the
social character of scientific inquiry that attempts to make sense of it as research
that is extended across events, environments, objects, and agents. My suggestion is
that responsibility for the form of ethical evaluations is therefore also distributed
across the system of agents within the network of practitioners.
So, what would a normative ethical decision-making process look like using
extended agency ethics? In pursuing research to find a suitable organism to be used
as a chassis for biofuel production, a team of researchers would use knowledge
acquired from the related areas of research outputs of projects focusing on cyano-
bacteria, algae, and metabolism. They may use this broad investigative approach to
narrow down the range of possible candidates for a chassis organism. They may
initially investigate the current algae research and the metabolic pathway of the
highly familiar, well-researched green algae, Chlamydomonas reinhardtii.
Knowledge of the successes and problems associated with the reengineering of
algae as a source of biofuel may lead to the choice of a cyanobacteria instead. Once
a chassis is selected, they may focus attention on the synthetic construction of path-
ways that overexpress alkane biosynthetic genes. Following this, they may begin the
task of generating stocks of the newly reengineered form of Synechocystic sp. In
order to plan the most efficient scale-up ventures, economists as well as microbiolo-
gists may be contracted. Once enough product is produced, they may outsource
some beta tests to chemical engineers for kinematic viscosity analysis of the
cyanobacteria-based biofuel product. They may request assistance in testing com-
bustibility from colleagues specializing in physical chemistry. Limitations on what
can be known and what can be done may come from the reciprocal knowledge
exchanged through these interactions as well as with the cultures and communities
potentially affected by the production of products. The normative ethical decisions
and projectable outcomes are obtained through and by these interactions. Knowledge
of the organism being used, the marketability of the product, the scale of produc-
tion, and the social and environmental impacts are all linked to the particular organ-
ism used. That is, the ethical evaluations of biofuel production varies depending on
the organism used, the scaling applied, the prospects for environmental controls,
expected social effects, communication of these, and the impact on local and world
economics.

[email protected]
58 C. Kendig

Extended agency ethics describes the integrated approach to ethical decision-

making as one that requires careful and critical consideration of integrated research
and technological activities in order to reasonably predict possible outcomes.
Ethical decisions concerning the potential use of Synechocystic- based biofuel is
not something that comes as purely either backward looking assessment or forward
speculation. The determination of what should be done is not something that can be
judged solely on the basis of weighing up consequences of action, nor on the basis
of rule- following—categorical or otherwise. Instead, ethical considerations—such
as the permissibility of the development of some new technology (e.g. cyanobacte-
rial- based biofuels)—are determined according to the current research, the experi-
ential data amassed, the practitioners’ knowledge-making activities, and the
potential for scaling up production of the biofuel products by industry. This means
that the locus of normative agency and intentionality is distributed across the activi-
ties of research groups, tools, and the development of products.
In this way, extended agency provides an externalist view of justification.
Justification for beliefs come not from the decision-making-inside-the-head version
of internalism (as some set of brain-bounded intuitionism or reflective perception).
Instead, a form of active externalism that distributes cognition socially and agency
across spatiotemporal research practices is suggested to explain the nature of scien-
tific inquiry and the acquisition of knowledge. The ability to form research ques-
tions and pursue this kind of research depends on the aims of the system of
practitioners within a particular research environment. As such, extended agency
radically revises the conditions under which beliefs are considered to be justified.
Justification can be on the basis of reflection on decision making as a form of mental
cognition, but according to extended agency, what is taken to be mental cognition is
not restricted to the brain but instead goes beyond the skull and can extend to tools,
practices, processes, and other researchers and social research groups—social cog-
nition. Both epistemic credit and ethical culpability are distributed notions that
extend beyond the individual human agent. If social cognition, the social aspect of
scientific practice, and the extended mind thesis are taken seriously, mental states
(usually restricted to the brain by internalist evidentialists) are extended not only
beyond the brain and skin of the individual but to include other spatiotemporally
distinct biological, technological, and socially extended entities. With this extended
cognition, a requisite extended agency and normative ethics based on this actively
externalist theory of evidence follows. In some meaningful sense, extended agency
can be understood to be a kind of role-based approach to ethics that demarcates
categories of valuation analogous to those epistemic categories (outlined in the first
half of this chapter). That is, to act in a role is to act according to a category of activ-
ity or to follow a model or prototypic way of acting. It may be profitably understood
as being akin to an Aristotelian notion of virtue or a context- driven valuation. The
intimacy of epistemic and ethical knowing is explicitly articulated by a number of
virtue ethicists (see in particular Swanton’s 2003: 249 “virtues as prototypes”).
Instead of understanding morality as being based on a set of rules, this approach
takes virtues to be frameworks. These frameworks are built from interactions and
in-practice experience that both shapes and is shaped by future interactions in the

[email protected]
Grounding Knowledge and Normative Valuation in Agent-Based Action and Scientific… 59

world (Swanton 2003: 279). Christine Swanton’s “virtues as prototypes” can be

seen as a more restricted version of my extended agency ethics. Whereas Swanton
limits knowledge of the world to agent-based interactions, I extend it further to
include agent-object based interactions. That is, I include interactions between
humans, but also between humans and their objects, technological artefacts, and
tools of investigation.
I hope my extension will help to build a bridge between philosophy of science
and sociology of science. The bridge can be understood in terms of a shared locus
of research: the study of interactions and associations between social actors. The
interactions between social actors—whether they be technological artefact, machine,
or human—consist of a system of valuing within these systems of scientific prac-
tices. The normative status of the interactions, objects, and human actors comes
from the valuing of practitioners, the scientific methods they choose, activities that
they participate in, the tools and measuring devices that they select in their field of
study to use, and the meaning attached to the information gained from those tools.
These all constitute sources of meaning within the use and interaction with them in
various ways. The aim of the extended agency approach outlined here has been to
begin to explain agent-object based interactions within scientific research groups. In
the next section, I suggest that these object-human interactions can be best under-
stood through a combined sociology and philosophy of science.

5.2 Bridging Philosophy of Science and Sociology of Science

In his pivotal paper, “Mixing humans and nonhumans together: the sociology of a
door-closer”, Bruno Latour, (writing under the pseudonym “Jim Johnson”), sug-
gests that studying the interaction between humans, machines, and tools should be
the remit of a more widely extended approach to the study of sociology of science.
He argues that if things we commonly refer to as a tool or a machine affect the way
we interact in the world, then they might also be considered social actors. Considering
them as such would blur the line that is often drawn between what is “purely technical”
and what is “purely social” (Latour 1988: 198). He shows by means of a series of
examples, (e.g. a hydraulic door closer and a red stop light), that some technological
objects prescribe the behaviours of humans who interact with these artefacts. He
focuses in particular on tools and technological objects that stand in for humans or
take on the work of humans, (e.g. the hydraulic door opener that takes on the job of
a human groom who opens the door for us, or a traffic light that takes on the job of
a police officer who signals that we should stop in traffic). Although they are not
human, the hydraulic door opener and the traffic light determine the norms of
behaviour considered appropriate when we encounter them and sets the normative
terms of interaction. That is, we interact with the non-human door-closer and the
traffic light in terms of the roles each occupies. Knowing the role these technologi-
cal objects play means that we also know what actions we should take in response
to them. They bring with them the norms for how to interact with them. That these

[email protected]
60 C. Kendig

technological artefacts might be informed by our interaction with a polite human

opening a door for us or a police officer who directs traffic in an area where no traf-
fic lights exist does not seem to count against the normative interaction present
when we drive up to a light and act according to certain norms. That the light is not
a person doesn’t mean we fail to understand that we should stop when it changes to
red. The object in some sense normatively frames our experiences with it. It inter-
acts with us in terms of how it interprets our role in the interaction. Latour calls the
normativity that arises from the machine’s presuppositions about the role of the
human user, “prescriptivism”.
Prescriptivism is whatever a scene presupposes from its transcribed11 actors and authors
(this is very much like ‘role expectation’ in sociology, except that it may be inscribed or
encoded in the machine). For instance, a Renaissance Italian painting is designed to be
viewed from a specific angle of view prescribed by the vanishing lines, exactly like a traffic
light expects that its users will watch it from the street and not sideways. In the same way
as they presuppose a user, traffic lights presuppose that there is someone who has regulated
the lights so that they have a regular rhythm (Latour 1988: 306).

Latour suggests we consider interactions with scientific objects and technologi-

cal artefacts (and not just people) to be social interactions. As such, he argues that
sociology should treat the material or technological objects we relate to as not just
of peripheral social interest but of direct sociological significance. Doing so requires
what Andrew Pickering calls “a decentering of the social relative to the material and
the conceptual, in terms of both objects of analysis and explanatory formats”
(Pickering 2005: 352). The current chapter attempts to follow Pickering’s approach
to decentered sociology by “exemplifying empirically and theoretically what a
decentered sociology can look like, in the hopes of encouraging others to follow”
(Pickering 2005: 353). In the above, I have attempted to provide a case study of how
the tools and assembly methods chosen by synthetic biology practitioners may be
themselves a sociological topic as well as a philosophical topic. They affect the
process of knowledge production as well as the normative judgements of practitio-
ners about the products produced within the collaborative social practice of research
itself. Sociological investigation of the material and conceptual tools that affect
knowledge production within labs may sheds light on how tools and decisions about
tool use within systems of practitioners are made. In this way, normative as well as
epistemic aspects of scientific research seem to require an understanding of social
relations and not just a philosophical explanation of knowledge production. A
decentered social theory and a decentered philosophical theory that focus on the
role of technological objects may provide the best account of the normative interac-
tions within collaborative groups of scientists.

11
Latour uses “transcription” and “inscription” to explain the transition from a less durable dele-
gated agent to perform an action to a more reliable one. For instance, “the replacement of a police-
man by a traffic-light” is an instance where the traffic light is delegated the work that was done by
a police officer. In Latour (1988), the focus is on descriptions of meaning that these actors play
within a particular semiotic script. How actors are defined and what is meant by their roles in a
particular scenario.

[email protected]
Grounding Knowledge and Normative Valuation in Agent-Based Action and Scientific… 61

6 Concluding Remarks

The account of both epistemological and ethical ineliminability of scientific prac-

tice developed here aims to provide non-reductive grounding to knowledge in
social cognition and within social enterprises. The possibility of critical dissent by
the individual practitioner against the protocols, types of reasoning, modes of
knowledge transmission, or generation of research questions affirmed by the com-
munity is what is required for knowledge. That requires the role of the practitioner
as individual agent to be something that is irreducible to the social group or system
of practice itself. Rather than a relationship of reduction of the individual to the
social, the individual practitioner as agent can been understood to be working
towards the goal of scientific knowledge and the performance of it using diverse
methods. The practitioner extends her knowledge by making connections with
other practitioners, making use of protocols endorsed by the community, using
reliable model organisms to make her hypotheses, critically responding to accepted
routes of knowledge generation, and networking her resources together in the per-
formance of scientific inquiries. As an agent, she brings these resources together
and in turn becomes a resource within the system of practice that she works within.
Investigating a practice-based route of acquiring knowledge within synthetic
biology presents an alternative way of exploring traditional metaphysical questions
in philosophy such as: What are the kinds of things synthetic biology produces?
What is the relationship between parts and wholes? It also introduces new questions
about what it is to be a part or indeed the property of partness itself within a new
field; questions not just of interest to metaphysicians of science. Instead of seeking
to understand the use of epistemological and ontological categories in practice from
the premise that their existence can be known a priori or contained within the theo-
retical framework of the discipline that uses them, this approach runs in the opposite
direction. It instead suggests that categories come into being in practice and from
these categories-in-use theoretical concepts, notions of causal directionality, func-
tional architectures, and normative valuation arise through the active engagement of
systems of practitioners.

Acknowledgements Research for this chapter was partially funded by the National Science
Foundation Division of Molecular and Cellular Biosciences (MCB), BIOMAPS: Modular
Programmed Evolution of Bacteria for Optimization of Metabolic Pathways, Grant No. MCB-
1329350, Amendment No. 001, Proposal No. MCB-1417799. Thanks to Todd Eckdahl, Jeff Poet,
Malcolm Campbell, and Laurie Heyer for sharing their insights and expertise in synthetic biology
with me during research for the project. Special thanks go to Phil Mullins for many lively discus-
sions about Polanyi and for encouragement in the early stages of writing this chapter. I am also
very grateful to Hauke Riesch, Brian Rappert, and Thomas Reydon for their feedback on earlier
versions of the manuscript.

[email protected]
62 C. Kendig

References

Brent, R. 2004. A Partnership Between Biology and Engineering. Nature Biotechnology 22:
1211–1214.
Calcott, B. 2014. Engineering and Evolvability. Biology and Philosophy 29 (3): 293–313.
Chang, H. 2004. Inventing Temperature: Measurement and Scientific Progress. New York: Oxford
University Press.
———. 2011. How Historical Experiments Can Improve Scientific Knowledge and Science
Education: The Cases of Boiling Water and Electrochemistry. Science & Education 20:
317–341.
———. 2012. Is water H2O? New York: Springer.
———. 2014. Units of Analysis in Philosophy of Science After the Practice Turn. In Science
After the Practice Turn in the Philosophy, History, and Social Studies of Science, ed. L. Soler,
S. Zwart, M. Lynch, and V. Israel-Jost. London: Routledge.
Chang, H. 2016. The rising of chemical natural kinds through epistemic iteration. In Natural kinds
and classification in scientific practice, ed. C. Kendig, 33–46. Abingdon/ New York: Routledge.
Clark, A. 1995. I Am John’s Brain. Journal of Consciousness Studies 2 (2): 144–148.
———. 2010. Supersizing the Mind: Embodiment, Action, and Cognitive Extension. Oxford:
Oxford University Press.
———. 1998. Being there: Putting brain, body, and world together again. Cambridge: MIT Press.
Clark, A., and D. Chalmers. 1998. The Extended Mind. Analysis 58: 7–19.
Dellomonaco, D., F. Fava, and R. Gonzolez. 2010. The Path to Next Generation Biofuels: Successes
and Challenges in the Era of Synthetic Biology. Microbial Cell Factories 9: 3. https://doi.
org/10.1186/1475-2859-9-3.
De Regt, H., S.  Leonelli, and K.  Eigner, eds. 2009. Scientific Understanding: A Philosophical
Perspective. Pittsburgh: University of Pittsburgh Press.
Dupré, J.  1993. The Disorder of Things: Metaphysical Foundations of the Disunity of Science.
Cambridge, MA: Harvard University Press.
———. 2006. Humans and Other Animals. Oxford: Clarendon Press.
Eckdahl, T.T., A.M.  Campbell, L.J.  Heyer, J.L.  Poet, D.N.  Blauch, N.L.  Snyder, et  al. 2015.
Programmed Evolution for Optimization of Orthogonal Metabolic Output in Bacteria. PLoS
One 10 (2): e0118322. https://doi.org/10.1371/journal.pone.0118322.
Endy, D. 2005. Foundations for Engineering Biology. Nature 438 (24): 449–453.
Erwin, D., and E.  Davidson. 2009. The Evolution of Hierarchical Gene Regulatory Networks.
Nature Reviews Genetics 10: 141–148.
Georgianna, R., and S.  Mayfield. 2012. Exploiting Diversity and Synthetic Biology for the
Production of Algal Biofuels. Nature 488: 329–335.
Gibson, D., J. Glass, C. Lartigue, V. Noskov, R.-Y. Chuang, M. Algire, et al. 2010. Creation of a
Bacterial Cell Controlled by a Chemically Synthesized Genome. Science 329 (5987): 52–56.
Goodwin, B., S.  Kauffman, and J.  Murray. 1993. Is Morphogenesis an Intrinsically Robust
Process? Journal of Theoretical Biology 163: 35–144.
Grene, M. 1966. The Knower and the Known. Berkeley: University of California Press.
———. 1985. Perception, Interpretation, and the Sciences: Toward a New Philosophy of Science.
In Evolution at a Crossroads, ed. D.J. Depew and B.H. Weber, 1–20. Cambridge, MA: MIT
Press.
Hacking, I. 1992. The Self-Vindication of the Laboratory Sciences. In Science as Practice and
Culture, ed. A. Pickering, 29–64. Chicago: University of Chicago Press.
———. 1995. The Looping Effects of Human Kinds. In Causal Cognition: A Multidisciplinary
Debate, ed. D. Sperber, D. Premack, and A.J. Premack, 351–394. New York: Clarendon Press.
Hornsby, J. 2004. Agency and Actions. In Agency and Action, ed. J. Hyman and H. Steward, 1–23.
Cambridge: Cambridge University Press.
Hylton, W. 2012. Craig Venter’s bugs might save the world. The New York Times, March 6.

[email protected]
Grounding Knowledge and Normative Valuation in Agent-Based Action and Scientific… 63

Keller, E.F. 2009. Knowledge as Making, Making as Knowing: The Many Lives of Synthetic
Biology. Biological Theory 4 (4): 333–339.
Kendig, C. 2014a. Towards a Multidimensional Metaconception of Species. Ratio 27 (2): 155–172.
———. 2014b. Synthetic Biology and Biofuels. In Encyclopedia of Food and Agricultural Ethics,
ed. P.B. Thompson and D.M. Kaplan, 1695–1703. New York: Springer.
———. 2016a. What Is Proof of Concept Research and How Does It Generate Epistemic and
Ethical Categories for Future Scientific Practice? Science and Engineering Ethics 22 (3): 735–
753. https://doi.org/10.1007/s11948-015-9654-0.
———. 2016b. Activities of kinding in scientific practice. In Natural kinds and classification in
scientific practice, ed. C. Kendig, 1–13. Abingdon/New York: Routledge.
———., ed. 2016c. Natural Kinds and Classification in Scientific Practice. Abingdon/New York:
Routledge.
Kendig, C., and T.T. Eckdahl. 2017. Reengineering metaphysics: Modularity, parthood, and evolv-
ability in metabolic engineering. Special issue: Ontologies of living beings (eds A.M. Ferner
and Thomas Pradeu) Philosophy, Theory, and Practice in Biology 9(8). https://doi.org/10.3998/
ptb.6959004.0009.008.
Kirschner, M., and J. Gerhart. 2005. The Plausibility of Life. New Haven: Yale University Press.
Knight, T. 2003. Idempotent Vector Design for Standard Assembly of Biobricks. MIT Synthetic
Biology Working Group.
Latour, B./Johnson, J. 1988. Mixing Humans and Nonhumans Together: The Sociology of a Door-
Closer. Social Problems 35 (3): 298–310.
Longino, H. 1990. Science as Social Knowledge: Values and Objectivity in Scientific Inquiry.
Princeton: Princeton University Press.
Lowe, E.J. 2009. A Survey of Metaphysics. Oxford: Oxford University Press.
Martin, V.J., D.J.  Pitera, S.T.  Withers, J.D.  Newman, and J.D.  Keasling. 2003. Engineering a
Mevalonate Pathway in Escherichia coli for Production of Terpenoids. Nature Biotechnology
21 (7): 796–802.
Minelli, A. 2003. The development of animal form. Cambridge: Cambridge University Press.
Morange, M. 2009a. Synthetic Biology: A Bridge Between Functional and Evolutionary Biology.
Biological Theory 4 (4): 368–377.
———. 2009b. A Critical Perspective on Synthetic Biology. HYLE 15 (1): 21–30.
Mullins, P. 2009. Polanyi on Agency and Some Links to MacMurray. Appraisal 7 (3): 11.
O’Malley, M. 2009. Making Knowledge in Synthetic Biology: Design Meets Kludge. Biological
Theory 4 (4): 378–389.
O’Malley, M., A.  Powell, J.  Davies, and J.  Calvert. 2008. Knowledge-Making Distinctions in
Synthetic Biology. BioEssays 30: 57–65.
Pickering, A. 2005. Decentering Sociology: Synthetic Dyes and Social Theory. Perspectives on
Science 13 (3): 352–405.
Polanyi, M. 1962. Personal Knowledge: Towards a Post-critical Philosophy. Chicago: University
of Chicago Press.
Rheinberger, H.-J. 2005. A Reply to David Bloor: Toward a Sociology of Epistemic Things.
Perspectives on Science 13: 406–410.
Rouse, J.  1996. Engaging Science: How to Understand Its Practices Philosophically. Ithaca/
London: Cornell University Press.
———. 2003. How Scientific Practices Matter. Chicago: University of Chicago Press.
Schlosser, G., and G.  Wagner, eds. 2004. Modularity in Development and Evolution. Chicago:
University of Chicago Press.
Schyfter, P. 2012. Technological Biology? Things and Kinds in Synthetic Biology. Biology and
Philosophy 27: 29–48.
Soler, L., ed. 2012. Characterizing the Robustness of Science: After the Practice Turn in Philosophy
of Science. Volume 292 Boston Studies in the Philosophy of Science. New York: Springer.
Soler, L., S. Zwart, M. Lynch, and V. Israel-Jost, eds. 2014. Science After the Practice Turn in the
Philosophy, History, and Social Studies of Science. London: Routledge.

[email protected]
64 C. Kendig

Sprinzak, D., and M.  Elowitz. 2005. Reconstruction of Genetic Circuits. Nature 438 (7067):
443–448.
Swanton, C. 2003. Virtue Ethics: A Pluralistic View. Oxford: Oxford University Press.
Wagner, G., M. Pavlicev, and J. Cheverud. 2007. The Road to Modularity. Nature Reviews Genetics
8: 921–931.
Wang, W., X.  Liu, and X.  Lu. 2013. Engineering Cyanobacteria to Improve Photosynthetic
Production of Alka(e)nes. Biotechnology for Biofuels 6: 69. http://www.biotechnologyforbio-
fuels.com/content/6/1/69. Accessed 1 Nov 2015.
West-Eberhard, M. 2003. Developmental Plasticity and Evolution. Oxford: Oxford University
Press.
———. 2005. Developmental plasticity and the origin of species differences. PNAS 102:
6543–6549.
Wilson, R. 2018. The Eugenic Mind Project, 99–140. Cambridge/London: MIT Press.

[email protected]
Hauke Riesch • Nathan Emmerich
Steven Wainwright
Editors

Philosophies and Sociologies

of Bioethics
Crossing the divides

[email protected]
Editors
Hauke Riesch Nathan Emmerich
Department of Sociology ANU Medical School
and Communications Australian National University
Brunel University Canberra, Australia
London, UK
The Institute of Ethics,
Dublin City University
Steven Wainwright
Dublin, Ireland
Department of Social & Political Sciences
Brunel University London School of History, Anthropology,
Uxbridge, Middlesex, UK Politics and Philosophy
Queen’s University Belfast
Belfast, UK

ISBN 978-3-319-92737-4 ISBN 978-3-319-92738-1 (eBook)

https://doi.org/10.1007/978-3-319-92738-1

Library of Congress Control Number: 2018948009

© Springer International Publishing AG, part of Springer Nature 2018

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the
editors give a warranty, express or implied, with respect to the material contained herein or for any errors
or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims
in published maps and institutional affiliations.

Printed on acid-free paper

This Springer imprint is published by the registered company Springer International Publishing AG part
of Springer Nature.
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

[email protected]