See the general introduction to plants and people
This article looks at the foundations of all classification systems before looking more closely at scientific classification in general and plant classification in particular.
Classification systems begin in our minds, but they must be explained and shared through the medium of language. This is where we must start.
Classification & taxonomy
Classification is here taken to mean the actual process of ordering or grouping objects using selection criteria intended to achieve some purpose or goal, while taxonomy is the study of the principles and procedures of classification. However, in practice these two terms are often used interchangeably. There is the further useful distinction of a category schema, the laying out of a complete classification.
Immanuel Kant, following the example of Aristotle had, in his transcendental analytic, classified modes of thought under the four categories of quantity, quality, relation, and mode. These were the a priori (innate) modes of thought – the logical principles – that order our mental experience. But the a priori characteristics of our experience are not confined to logical principles.
For example, we do not experience everything, all at once. To cope with the flood of incoming sensory information our brains segregate the world into meaningful objects, focus on a limited range of these objects at any given time, often classifying and ranking them into meaningful groups according to conscious and unconscious priorities. The outcome is that we exist in a world that is not a chaotic maelstrom of competing sensations, but a unified, coherent, and meaningful experience.
Clearly the ability to order the world like this is an evolved and innate biological characteristic, and therefore part of our human nature.
At present my attention has settled on writing this article, but soon it may be diverted by a visitor or the desire for a cup of coffee as my mind responds to prioritizing factors in my internal states and external circumstances. But when these diversions are satisfied my focus returns to the computer screen and the selection (prioritization, classification) of possible words that I can use, their order within a sentence, and so on. Clearly, at any given moment, we are experiencing multiple conscious and unconscious prioritizing processes as we think and feel.
Our mental activity includes at least four innate and interconnected processes. We know these processes are hard-wired in us because they occur in all cultures and are, indeed, essential preconditions for operating effectively in the world. These processes are described elsewhere under the heading of mental processing, but are summarized here as segregation, focus, classification, and ranking.
Segregation structures our world into meaningful representational units of experience, both those of cognition (concepts), and those of perception (percepts).
Our minds not only fragment our experience into meaningful categories, they also focus our attention on a restricted range of these categories so that, at any given time, our attention is divided into a foreground and background.
Classification arranges the objects of our attention into groups according to grouping criteria
Formation of groups, as a classification, may be based indifferently on similarities and differences (as when I distinguish between tables, chairs, and shelves in my room) but, more often, groups are ranked in some way relative to one-another (valued, prioritized, specially selected) based on our needs, desires, purposes, and reasons – as when I go shopping. The ranking or prioritizing of objects relative to one-another I call rank value.
We frame our comprehension of the world in terms of units or objects of cognition (concepts) and perception (percepts). For simplicity we can call all units of experience categories, and it is these categories that can be variously grouped and prioritized according to grouping or selection criteria.
Classification is classification of objects of any kind – whether they be concrete or abstract, and whether we treat them as things in the world or concepts in our heads. The use of the words ‘categories’ and ‘categorization’ (= classification) is in line with late 20th century psychological research into the mental process of ordering and differentiation. This is a core mental process that is a crucial part of both the organization of knowledge within the cumulative process of collective social learning, and the mundane way we arrange our day-to-day lives from one moment to the next. We could not survive without it. ‘Categorization’ is a portmanteau term covering all of this processing while ‘classification’ is a word that we tend to use for categorization that we take seriously in a shared social context. Viewed this way, classification is a sub-category of categorization.
Groups formed by categorization become further categories in a complex system of interrelated groups and sub-groups, parts and wholes, all available for more mental processing.
Among this multitude of categories are those denoting physical objects in the world. These can be referred to as ‘kinds’ and one subset of these kinds is that of the living organisms that constitute the community of life.
Scientific categories aim for maximum economy and precision in their expression and this begins with the metaphysical observation that we can consider the world of scientific objects in toto as consisting of kinds, their properties, and relations.
Kinds of categories
Categorization and classification (essentially synonymous) have multidisciplinary application and therefore conflicting terminologies. In taxonomy (the study of classification), any unit of classification is referred to as a taxon (pl. taxa). In theory this word should have wide application, but in practice it is largely confined to units of biological classification. Cognitive psychologists refer to category-specific mental representations as concepts. In general parlance we have the words sets, groups, divisions, classes, domains, and so forth. Biologists speak of clades.
Categorization allow us to organize and explain our experience and has long been studied by philosophers who desire categories that are both clear and distinct. Plato wanted to establish categories of the external world by ‘carving nature at the joints’, by which he meant dividing the physical world into units that we believe exist in nature rather than in our minds.
Aristotle described what we now call ‘classical’ categories which had defining features (selection criteria) that were both necessary and sufficient. These defining features were then known as ‘essences’, and their use was called ‘essentialism’. Such categories were thus clearly defined, mutually exclusive, and collectively exhaustive, which did indeed provide a grounding for categorization in clear and distinct ideas. One example would be the periodic table of elements defined by atomic number. Though gold has many properties, scientist are generally agreed that a definition of gold as the element having the atomic number 79 is both necessary and sufficient. Unfortunately, such clarity is found more in logic and mathematics than in nature.
Most categories applied to natural objects (and the meanings of most words), are more blurry; they are graded at their boundaries. As the philosopher Ludwig Wittgenstein pointed out, the idea of a ‘game’ is not a classic category but one that is fuzzy at its boundaries, grading into other ideas. What we call ‘games’ share a ‘family resemblance’, but no ‘essential’ features. Awareness of the wide application of fuzzy categories has given rise to new conceptual clustering tools like those developed in fuzzy mathematics (fuzzy logic and fuzzy set theory which developed in the 1960s) in which objects may belong to one or more groups but to varying degrees. Psychologists studying ‘category learning’ are investigating the possible ways our brains compare categories and establish group membership.
Categories in a taxonomy are related to one-another via degrees of group inclusion with the most inclusive also being the most abstract and abstract while the lowest levels are the least inclusive and abstract and with the greatest degree of specificity and within-category similarity. Passing from greater to lesser generality and inclusivity: we have, for example . . . tree, eucalyptus tree, coolabah tree (classifications using analysis and synthesis)
We do not think in words (we think in ‘mentalese‘) but to share our cogitations with others we must translate mentalese into a spoken and written language.
We may think of the way we structure language by invoking the metaphor used by cognitive scientist and linguist Steven Pinker who refers to it as ‘the web, the string, and the tree’.
We have in our minds a web of interrelated categories. We organize these ideas into a linear sequence (spoken or written) – a string – according to the rules of syntax which, together with the rules of word formation comprise its grammar. Grammar is our species solution to getting complicated thoughts from one head to another’. The words are then arranged into phrases that are nested within sentences in the form of an inverted hierarchical tree.
‘Syntax then is an app that uses a tree of phrases to translate a web of thoughts into a string of words.’ The listener then works backwards, fitting them into a tree and recovering the links between the associated concepts.’
So, we structure language like a nested hierarchy that is constructed in a boxes-within-boxes way with sentences containing substructures (phrases) that form an inverted tree if presented pictorially on the page.
Language provides us with a symbolic representation of the world. The concepts and categories that make up our language are, like the universe itself, linked into a web of interconnected knowledge. We need to ensure that the categories we use to organize this knowledge are the best-for-purpose that we can possibly produce. This is the task of science.
Knowledge grows cumulatively and to make an impact in any field of study a student must add to the fine detail of their particular discipline by refining existing categories and adding new ones. As students we can become enmeshed in our own particular region of the knowledge web. Proceeding by analysis and the desire to understand the fine detail we do indeed ‘learn more and more about less and less’ and as individuals it becomes increasingly difficult to establish a synoptic view of the intellectual landscape – something we desperately need to reassure ourselves that effort and resources are being directed towards the areas of greatest need.
Parse Tree as a Nested Hierarchy
Courtesy Wikimedia Commons – Tjo3ya – Accessed 3 January 2021
The classifications we create are decision-making tools. We devise a classification for a purpose and the best classifications are those that fulfil, in the most efficient way, the goals that we want to achieve.
Thus, classification allows us to impose order by organizing categories into groups according to selection criteria chosen for an objective or purpose. When groups are then prioritized or ranking in relation to one-another, this is called rank-value.
Selection criteria (characters) are those features that are chosen to filter the purpose of the classification. By narrowing down possible outcomes this increases the precision of the result.
Categorization gathers complexity with the increasing number of selection criteria. A simple alphabetical list is an ordering based on one selection criterion, letter of the alphabet. Organisms may be classified using selection criteria in such large numbers that a computer is required to build the classification system.
Units of classification
Classification proceeds by the grouping of categories – and any category may serve as a unit of classification.
A meta-question immediately arises: ‘Are there more useful way of classifying all these categories?’
Suffice it to say that this (metaphysical) question ‘How should we classify everything that there is?’ has been an open question throughout the history of philosophy. We will not be able to answer it here: but, for a longer discussion, I invite you to read the article on reality.
For the time-being we can note that one way of dividing up the world, of grouping or classifying its contents, was suggested by that great metaphysician Aristotle who suggested that ‘everything’ consists of objects, their properties, and relations. This is a valuable distinction that will soon become evident.
There is, however, one sense in which this question has already been answered: ‘It depends on the purpose of the classification’.
Regardless, it will help to narrow the field of categories down (removing a host of mental categories) by confining ourselves to those categories that we take to represent objects in the physical world. So, for our purposes, the word ‘kind’ will refer to categories found useful in the field of science.
More specifically we are concerned with kinds that are found in biology, and not just any kinds, but kinds of organisms.
Individual organisms are variously grouped into categories called species, genus, family etc. The specialist term for one of these is a taxon (i.e. taxonomic group, pl. taxa). Though the term taxon can denote any unit within any classification, it is usually applied to aggregates of living organisms.
Foundations of taxonomy
Before moving on to plant classification it is now possible to draw up some conclusions about taxonomy in general.
Classification begins with the pre-conscious structuring of the objects of our experience (categories) using selection criteria directed towards some end or purpose. This is the way that we intuitively relate to the world – the way that we innately order the world before conscious deliberation begins. Four interrelated aspects of this innate mental processing can be usefully distinguished: segregation, focus, classification, and ranking. This determines much of our human umwelt, our sense of reality.
Not all classification is innate and intuitive – it can occur with varying degrees of consciousness, and with some aspects being unconscious and others conscious. At one extreme might be the way our eyes move intuitively towards the source of a noise but, much more complex, there is the both conscious and intuitive prioritizing in importance of the objects around me when I am driving. The solving of a mathematical problem seems a fully conscious process. When classification is fully conscious it becomes part of the process we refer to as ‘reason’.
Clearly the success of any classification depends on the effectiveness of the selection criteria chosen to achieve the purpose for which the classification was devised.
It is now obvious that plants can be classified in countless ways, and this begs the question as to why we treat the scientific classification of plants with such respect, or as being in some way special? So, what are the selection criteria for the scientific classification of plants – not the individual plants but the classification system itself?
Systems of classification
It is now time to look more closely at the actual process of grouping objects. Are there different kinds of classification – different ways of organizing units into groups?
Special & general purpose
Before looking at classification systems in general, a distinction should be made between general- and special-purpose categorization/classification.
Most of our mental categorization is of no interest to others – like ordering my daily activities, shopping list, what I do next etc.). These can be called general purpose categorization. But sometimes categorization serves an important social service – like scientific classifications of many kinds, transport timetables, and so on. These latter require careful deliberation that might test our collective learning to its limit.
Many classification systems exhibit some kind of hierarchy, that is, they rank or prioritize their objects so that related objects are arranged in metaphorical levels, like the rungs of a ladder with a ‘top’ and ‘bottom’.
In a strict nested hierarchy the groups are contained within other groups like a Russian doll, which contains another doll, which contains another doll and so on.
A scientific example would be the Linnaean biological classification of organisms into orders which contain families which, contain genera, which contain species . . . etc. Knowing that a plant has the species name Lactuca sativa (Lettuce), connects us to a vast store of associated historical and biological information. An example of how this kind of plant classification works at high levels would be the single category ‘seed plants’ which is then divided into the groups ‘flower-bearing plants’ and ‘cone-bearing plants’. The flower-bearing plants can then be divided into those which have one seed leaf (monocotyledons) and those which have two seed leaves (dicotyledons) . . . and so on down the hierarchy of ranks.
Strict hierarchies like this exhibit several important properties:
Inclusivity – they are progressively more inclusive as ranks go from bottom to top
Exclusivity – an item in a strict hierarchy can only belong to one group at a particular ‘level’ or rank.
Transitivity – the properties that define the objects at higher ranks pass on to the lower ranks
Clear boundaries – the properties defining group membership at a particular rank must be both necessary and sufficient
An example of transitivity would be that all vertebrate animals (Vertebrata) have a vertebral column no matter how they are subsequently subdivided; sub-groups though resembling one-another by sharing the properties of ‘higher’ groups nevertheless differ in the properties that uniquely define them. So, Homo sapiens as a species shares the property of being human as expressed in the genus name Homo, but it differs from other species of human by being H. sapiens, the ‘Wise human’. There is thus a clever association and distinction) – the binomial (a name consisting of two words) expresses both similarity and difference at the same time
An example of exclusivity would be that Homo sapiens cannot be, at the same time, Homo heidelbergensis.
Strict nested hierarchies require clarity about similarities and differences between groups. Hierarchical definitions express both affinity and distinction in an economical way.
Transitivity allows the hierarchical structure to build up information that is useful for prediction, while inclusivity (the containment of groups within groups) demonstrates the evolutionary principle of descent with modification.
The formal structure of nested hierarchies works well for items that can be clearly defined, because they have unambiguous group boundaries. However, in daily life we use many categories with indistinct boundaries so the demand for mutual exclusivity cannot be met satisfactorily. In such cases the members of groups at a particular rank sharing family resemblance rather than fulfilling strict necessary and sufficient conditions – and some items at a particular rank may share closer resemblance than others.
In addition, limited knowledge can diminish the power and reliability of definitions and grouping items using more than one criterion can become unwieldy (say classifying plants based on both flower structure and fruit structure).
The effectiveness of a hierarchy in achieving its purpose will only be as good as the knowledge used in its construction and lack of clarity about items and their definition diminishes the power of transitivity.
Difficulties in creating clear categories in science are well known, for example, in the classification of rapidly-changing viruses or in defining the nature of particular smells. Descent with modification can itself lead to graded transitions rather than necessary and sufficient groupings (essentialism) as we try to impose logical order on graded nature.
Another kind of hierarchy is exemplified by military rank. This hierarchy has clear distinctions between groups (like private, sergeant and lieutenant) but there is minimal transitivity between the groups: that is, the characteristics shared by a private and a sergeant are somewhat obscure.
So, trees lack inclusivity: though a species is a member of a genus which is a member of a family – a private is not a subdivision or kind of Sergeant who is, in turn, a kind of Lieutenant. A tree indicates a chain of command but does not include a clear definition of the nature of the authority.
Tree – a nested hierarchy of Army Officers.
Trees lack inclusivity
Courtesy Wikimedia Commons – Totobaggins – Accessed 5 January 2021
A further kind of hierarchy is the partitive hierarchy. A partitive hierarchy might arise, for example, when we subdivide a geographic region. Melbourne is a part of the state of Victoria, which is a part of Australia. This is an inclusive hierarchy (though Victoria is not a kind of Australia it is a part of it) where certain properties are shared.
Partitive hierarchies share, with nested hierarchies, the characteristic of passing from the general (the most inclusive domain) to progressively smaller or less inclusive parts. But note that a simple confusion can arise here. Objects that share the characteristic of just being part of something can be quite different in general character: an apple core, a train ticket and bottle may all be parts of the waste bin while, in a nested hierarchy of animals we know that those animals that are vertebrates will share certain similarities.
Trees require a knowledge of the characteristics of the items being classified. The structure of the tree will be determined by the nature of the relationship between the parts: part-whole; cause-effect; process-product; start-end etc.
Trees organize categories in a way that defines how they are related and/or the degree to which they are related (spatially, metaphorically), and/or the relative frequency of items within a particular category. However, a tree is constrained by the order in which distinctions are drawn and this may require subsequent modification.
In a hierarchy the information flows (metaphorically) not only ‘upwards’ and ‘downwards’ between levels, but ‘laterally’ between items at the same level.
In trees the ‘lateral’ categories might contain very different objects so they tend to be strong along only one dimension of interest and are not so effective at representing multidirectional complex relationships. As with hierarchies the choice of key defining characteristics can be a matter of dispute and trees allow only partial inference.
It is important to be aware of the strengths and weaknesses of these forms of classification. Each method has its own goals, structural properties, strengths, and weaknesses.
New knowledge might initiate the formation of new classifications, but classifications may themselves contribute to knowledge generation. This occurred, for example, with the prediction of new and, as yet, undiscovered elements and their properties following the construction of the Periodic Table, and the new classifications and categorisation resulting from new scientific technology, as with: carbon dating, DNA analysis, remote sensing, radio-astronomy, the crystalline structure of gems rather than their hardness and so forth.
The goal of these classifications, like that of all science, is to maximise representational power.
Classification & progress
One advantage of viewing science as the progressive refinement of its categories of understanding and explanation is that it conveys, in a simple way, how science (and indeed any study) can be made progressively more efficient and effective. The advance of science is not just about discoveries.
Not all classifications are progressive. Going shopping requires all kinds of mental ordering of our purchases, but we are not making scientific progress. The advantage of scientific categories are that they are shared (not individual or personal) categories whose refinement is a matter of broad agreement within the community of scientists.
There have been, throughout history, obvious ways of refining scientific categories – as in the discrimination of ever smaller particles of matter – and ever larger expanses of space. Biologically we have achieved an ever finer resolution of the numbers of taxa, their geographic distribution, and evolutionary relationships of the organisms of the world. Indeed the resolution of categories has has been so rapid that the number of individual disciplines has rocketed.
Progressive scientific classification has been so successful that we might assume it is in some way a superior form of classification surmounting all others since it is more accurate, reliable, and predictive – of all systems of classification, scientific classification is the best. But this could be a mistake, and a look at scientific plant classification will tell us why.
Philosophers speak of science as dealing with natural kinds, which are categories that reflect the structure of the natural world rather than the interests and actions of human beings. Part of the success of science we associate with this achievement when the classifications and taxonomies employed by science correspond to the real kinds in nature. The existence of these ‘real’ and independent kinds of things is then used to guide and justify inferences and practices.
It is therefore taken that one strength of science is that it uses categories that relate to the external physical world (natural kinds), rather than the imaginative categories that our minds are wont to create. The ‘kinds’ of science (not just taxa, but all scientific categories) have been subjected to intensive intellectual scrutiny and, as a result, we are able to predict (and therefore manage), with a high degree of accuracy, the events that occur in this world outside our minds.
Today, plant practitioners give special credence to scientific classification because it lives up to Theophrastus’s ideal of objectivity by reflecting, as far as possible, the way plants are related to one-another – not to human utility or the unrelated organisms and the inorganic constituents of the environments where they are found. A scientific classification minimizes the influence of human interest and subjectivity.
Scientific plant classification
Just as our minds need to operate with discrete mental units (categories), so we need to establish basic units of the plant world, preferably with categories that relate clearly to physical objects. This allows us to manipulate (classify in various ways) the basic units of the plant world. The scientific classification of plants is an ontology, a statement of ‘what there is’, plant-wise, in terms of the kinds of plants that there are, and their relationships. Significantly, though these units (taxa) are treated intellectually as discrete, their grouping is under constant scrutiny and improvement.
Evolution of scientific plant taxonomy
If we view scientific plant classifications as those groupings of plants that have been adopted by the plant experts of their day – plant practitioners and academics of various kinds – then we can see a historical path of development in the selection criteria used for plant classifications in general. This change, it would appear, reflects both a predictable path of collective learning as well as the changing interests and perspectives of the day.
First came plant utility, the use of plants for food, medicine, structural materials and so on. With plants the common factor in so much of importance to human life and activity, the advantages of distinguishing one kind of plant from another were obvious. Though it was plant properties that were of primary interest (their key selection criteria) there were obvious benefits in being more proficient at distinguishing one kind of plant from another. In very general terms, the emphasis of plant practitioners had moved from the relationship of plants to people, to the relationship of one plant to another.
The need for plant ‘units’ served the same purpose as the need for units of perception, cognition, meaning, and language. It provided the means for linking, manipulating, and improving, our understanding of the relations between plants and the other categories our our understanding – the web of ideas that is part of our collective learning.
The study of plants had emerged from more general human activity in a process of naming, description, and classification – a phase of plant study appropriately known as ‘botany’ which emerged as an academic discipline in the mid-16th century.
Before Darwin, in the mid-19th century, botanists had noticed that plants shared similarities and differences and that they therefore shared different degrees of affinity with one-another. Even so, most people believed that god had created each species de novo as separate and immutable. The great Linnaeus believed that his work on plant classification in the mid-18th century was the revelation of plant order placed on earth by god.
In the mid-19th century Charles Darwin changed all this when he proposed that the entire community of life had arisen from a common ancestor. Modern classifications therefore seek to reflect, not only static similarities and differences, but also the dynamic changes that have occurred in the course of evolutionary history.
At the time of Darwin, plant study was beginning to broaden out once again as it linked once again to the human community. Agricultural science emerged stimulated by discoveries in chemistry and the new plant physiology, while plant description was putting a new emphasis on plant spatial distribution as new plants were drawn in from geographically distant colonial empires, and scientists like Alexander von Humboldt were initiating new scientific categories of plant distribution across the world and in relation to other environmental factors. The categories of ecology were now open to the process of collective scientific refinement.
The study of plants had entailed three core relationships: between plants and humans, plants and plants, and plants and their environment.
A giant leap in the study of plants was taken in ancient Greece. This, so far as we can tell, was the true dawn of plant science when Theophrastus, who followed Aristotle as Head of the Lyceum in ancient Athens, recognized that, for all prior history, people had investigated plants from a human perspective, classifying and studying them according to their uses . . . as food, medicine, materials, and so on. This was the study of plants as a human resource, and classifications therefore employed utilitarian selection criteria.
Aristotle, who founded the Lyceum, famously declared:
‘How tiresome it is to keep asking of natural things – what is its use? . . . Once we have got what we need to survive, we should turn our attention to understanding nature in terms of its own ends and goods‘.
The great Aristotle-Theophrastus insight, then, was that plants could be studied for their own sake . . . their structure, function, relationship to one-another, their environments and more . . . without regard to human interests.
So, up to the time of Greek science this perspective reflected the relationship between plants and humans, mostly in relation to their medicinal properties. Though it seems likely that plant use for foods would take precedence, it was the medicinal properties that comprised the specialized ‘scientific’ knowledge of respected community plant practitioners – the precursor ‘educated academics’ of old, be they the shaman, priest, or scribe.
At Theophrastus’s Lyceum in ancient Greece it became the relationship between plants and plants that became the key criterion of classification.
But this was a brief intellectual flame that was quickly extinguished for over 1000 years as the basis of plant study returned to the utilitarian relationship of plants and humans; classifications that characterized the post-classical world of materia medica, apothecaries, monasteries, and herbals that existed across both Europe and Asia.
Detachment from human concerns is often regarded as an indication of sound science. But it would be nearly 1600 years after Theophrastus before the study of plants in relation to plants would return. Theophrastus had studied plants in many kinds of relationship but from the Renaissance to the at least the mid 19th century it would be the naming, classification and description of plant kinds that would dominate the study of plants. The study of plant morphology and identification exemplified this scientific approach to plants but it was still generally also allied to utilitarian medicinal properties. Though questions about plant function and interaction with the environment were asked, it would be some time before carefully reasoned scientific answers would be given.
So, the scientific classification of plants was devised (in line with the original intentions of Theophrastus) to study the similarities and differences between plants themselves so as to organize them as efficiently as possible into groups based on their physical characteristics.
The purpose or goal of scientific plant classification is therefore to ascertain the number, kinds, and relationships between the world’s plants. This scientific endeavour was at its most pressing during the Ages of Discovery and Enlightenment when Europeans were mapping the physical boundaries of the world’s continents, no doubt an incentive to accelerate the inventory of the world’s biota. At that time the economic and cultural potential of the world’s unknown plant resource was of foremost concern. Plants like spices had made men fortunes and given some countries power over others.
After the Renaissance, as emphasis on comparative morphology increased, there was a return once again to the ancient Greek classification based on the relationship between plants and plants. This is the approach that has, by-and-large, continued until the present day although a profoundly different perspective emerged with the publication of On the Origin of Species . . .. Before Darwin, grouping were made based on morphological similarities and differences. Though this continued after Darwin, this took on a different meaning. Formerly similarities and differences were mostly treated as differences that arose when god created new and immutable species. This was essentially the view of plant classifications greatest exponent of these times, Carl Linnaeus. Humans, in their taxonomy, were organizing god’s creation.
Darwin presented a totally new account of the community of life as descent, with modification, from a common ancestor. The purpose of scientific classification then is to provide insight into the relationships of all organisms as they have evolved from a common ancestor. Closely related organisms possess similar properties and this becomes valuable knowledge in the establishment of predictability. Botanists now make a very clear distinction between artificial plant classifications such as those based on flower colour or edibility (a subjective classification) and natural classifications based on flower structures that reveal the evolutionary relationships of plants (much more objective).
There is a third kind of plant classification, essentially that introduced by Alexander von Humboldt. He selected as a key criterion of purpose and perspective, the relationship between plants and nature. This form of plant classification has been largely ignored or, perhaps better expressed as distinctly subsidiary and of little consequence for classifications based on plant/plant comparisons. In my view, this has had a profoundly unfortunate consequence for the history of botanical science as so much money and resources has been poured into plant systematics that could have been directed towards plant ecology and the attempt to better ascertain the scientific relationship between humans and the world’s vegetation.
The purpose of a scientific plant classification, then, is to mirror as accurately as possible what exists in the world. This, then, facilitates prediction and therefore the understanding and management of the objects it contains.
It turns out that the classification systems used in science are not unique to science – they are universal: they are the way we organize objects in our minds, and the way we organize them in our language. What then are the various ways of classifying things – what are their goals, structural properties, strengths and weaknesses?
Theophrastus’s division of plants into ‘herbs, shrubs, and trees’ was a practical classification that persisted into the 18th century. Today formal scientific classification groups plants in relation to one-another in terms of their presumed evolutionary (phylogenetic) history, sometimes known as phylogenetic systematics. The one-time morphological characters observable with the naked eye have been supplemented by microscopic and chemical-genetic information that is analysed using complex computer algoriths. Classification that concentrates on genetic characters is referred to as molecular systematics.
It is virtually impossible today to define what we mean by ‘science‘, but one devious way would be to say that it is the refinement of the classification of all those categories that we use in the field of activity that we refer to as ‘science’. So, for example, science is involved with the classification of physical objects like animals and plants. But science is much more than the classification of physical objects it also deals with more mental categories like names, laws, hypotheses, definitions, principles, theories and so on, and part of the scientific process is the refinement or improvement of these categories as well.
There are, then, three major phases of plant study: first, the close observation of plants in the general realm of human activity but especially through the specialist skills of medicine men and priests who had both knowledge and skills in the plant properties and their relationship to the spiritual world; second, a phase of plant inventory and description best described as botany; third, the expansion of studies beyond morphological and anatomical description into physiology and ecology as plant science.
Though we might seem to have clear and distinct concepts of what constitutes an individual plant, the idea of a species is controversial. Are species ‘real’ entities existing in nature, or are they creations of the human mind – or a bit of both?
Scientific plant classification
So far we have found that plants may be grouped in countless ways and that the outcome, the resultant classification, depends of two major factors: the selection criteria on which the classification is based, and the particular system of classification that we use. It was suggested that the range of topics with the potential to become subjects of serious plant study fall into three groups. First, those that involved the relationship between plants and people and among which would be studies of plants as food, medicine, and the materials of potential use in daily life, like building materials, dyes, resins and so on. Second, for its own intrinsic interest, and to assist any other studies involving plants, there was the need to establish an inventory of the plant world as, essentially a comparative study of plants in relationship to one-another. Thirdly, there were questions to be answered about the way plants functioned (physiology), their spatial distribution across the world, and their relationship to other organisms and their environments. Some work on all these topics was, of course, being carried out from the earliest days. However, it might be concluded, in retrospect, that it is unfortunate so much time was spent by the botanical scientific community on descriptive botany when there were, in addition, so many other important and challenging matters to be addressed.
From about the mid-15th century into the mid-19th century the special study of plants was narrowed down to the task of establishing kinds, what I refer to as ‘botany’. It is this narrowed down understanding of plant classification (and what is commonly understood by the expression ‘plant classification’) as the delineation of plant kinds, that will now be briefly addressed. Text-books on this mode of plant classification are part of any biological course so I shall be brief.
To take on the universal character of a science, botany needed a community of communicating botanists with a commonly accepted toolbox of ideas. Before classification could proceed there needed to be an agreed set of basic biological units of classification (taxa), along with an agreed terminology for their structures, along with a method of inventory and description that included a system of nomenclature and classification. In the modern era it took over 200 years to develop these scholastic tools whose integration into an internationally-acceptable system was the great achievement of Carl Linnaeus. It was Linnaeus’s forging of this common scientific ground that constituted his great achievement, not just his reinforcement of the system of binomial nomenclature.
In Linnaeus’s day the relationship was judged in terms of the overall similarities and differences that would indicate affinities. Classifications based on these characters were called natural classifications. However, sometimes it was easier to group plants according to simple and obvious characters like flower colour, and these were known as artificial classifications. Linnaeus himself used a system of artificial classification based on the number of stamens and styles in the flower.
When Darwin published On the Origin of Species . . . in 1859, the meaning of ‘natural’ changed. Biological classifications became progressively focused on hypothetical evolutionary relationships, not just superficial resemblance. As it happened, those organisms with similar characteristics were usually closely related evolutionarily but not always, as in the case of parallel evolution. The selection criteria now became more focused on characters indicating common descent (hypothetical evolutionary trees now calculated by sophisticated computer programs using a wide range of characters) rather than simple similarity and difference. As it happened, descent with modification fitted in very well with the boxes-within-boxes strict nested hierarchy form of classification that had been used by Linnaeus.
Of all the various projects that might be subsumed under the category ‘study of plants’, up to the mid-19th century the key endeavour remained the process of inventory- the naming, classification and description of plants. This project has, unfortunately, dominated beyond all reason the study of plants and persisted into the present day. This was partly a consequence the preoccupation of the best botanical minds in Britain with recording of plants within its empire. But, just as botany reluctantly shook itself from the clutches of medicine, to descriptive botany has been slow to extricate itself from the wider relationships of plants, humans, and the environment. German botany began the fragmentation of plant study with major breakthroughs in plant physiology that were of consequence to agriculture, bringing plant study back into the world while the new ecology, the role role of plants within the environment as a whole, had also emerged in Germany with the work of Alexander von Humboldt, amplified by Ernst Haeckel.
We need to know the units that underlie any study, but the course of historyand its environmental concerns long ago outstripped the demand for taxonomy.
However, on the taxonomic front increasing numbers of characters were used to map, with ever finer resolution, the similarities and differences that existed between different kinds of plants.
The advent of computers and gene technology has subsequently given plant classification greater discriminatory precision, and therefore increased predictive power – all part of the progressive accumulation of plant collective learning.
While the selection criteria use to generate general plant classifications depended, very broadly, on three kinds of relationship: that between plants and people (by far the most common way that we have always grouped plants), plants and plants (as adopted by botanists and plant scientists), plants and the environment (as emerged with the scientific adoption of ecological ideas.
In the 19th century the project of plant scientists moved beyond the delineation and description of taxa into other fields of learning.
It remains to acknowledge that as a consequence of the historical development of of the study of plants of the many scientific plant classifications that could have been, the one we speak of as ‘plant classification’ is the one that organizes plant kinds. It is this path that we follow now, from its beginnings.
From about 300 BCE to about the middle of the 18th century plants were grouped together according to their habit as trees, shrubs, herbs and so on. The purpose of the classification was to group plants according to simple and obvious characteristics of the plants themselves rather than any particular utility they might have for humans.
During the 18th century with a more universal terminology that included a broad range of plant parts, the selection of possible characters became more numerous as plants were further divided using a system of artificial classification which used just a few simple and convenient characters that facilitated plant identification to a finer resolution of taxa, as the number of plant species rose from about 1000 at the end of the classical era, to around 10,000 in Linnaeus’s day. The best known of these artificial systems was the ‘sexual system’ of Linnaeus which grouped plants according to the numbers of sexual parts in the flower. The selection criteria, though based on differences between plants themselves, were nevertheless practical characters of human convenience.
Natural general-purpose classification
As plant knowledge rapidly increased during the 18th century this artificial system was further refined using a much larger set of characters. Emphasis was now placed on overall similarities and differences between the plants, rather than characters that were simple and practical to manipulate. This became known as natural classification because it grouped plants according to the way they seemed to be related in nature. Botanists were no longer imposing order on the plant world by using an arbitrary human choice of convenient characteristics, instead through the close examination of many characteristics they were discovering the order that existed within nature itself. Most botanists, like Linnaeus, believing that this was the ‘natural order’ established by god which they were gradually revealing.
The comparison of similarities and differences was sometimes referred to as a general-purpose (phenetic) classification since as many characters as possible were taken into account with no character regarded as more important than any other.
Flowering plant phylogeny as at 2020
Including the currently recognized 435 plant families
The colours of the branches and the outer circles represent the major angiosperm clades (branches of the plant evolutionary tree) as indicated in the caption (ANA, Amborellales + Nymphaeales + Austrobaileyales). The illustrations around the circumference show 32 plant families and their position in the tree.
The flowering plants (angiosperms) arose in the Early Cretaceous about 145–100 million years ago, supplanting the earlier ferns and conifers (pteridosperms and gymnosperms). Palaeobotanical evidence indicates that this replacement was gradual, flowering plants not becoming dominant until the Palaeocene about 66–56 million years ago. The timing, geographic sequence, and plant composition of this diversification remains uncertain. There appear to have been substantial time lags, mostly around 37–56 million years, between the origin of families (stem age) and the diversification leading to extant species (crown ages) across the entire angiosperm tree of life. Families with the shortest lags occur mostly in temperate and arid biomes compared with tropical biomes. The ecological expansion of existing flowering plants, it seems, occurred long after their phylogenetic diversity originated during the Cretaceous Terrestrial Revolution.
Courtesy public domain license. Adapted from the originals provided by S.M., the Peter H. Raven Library/Missouri Botanical Garden and the Mertz Library/New York Botanical Garden
Phylogenetic special-purpose classification
After the mid-19th century and the publication of Darwin’s theory of evolution by natural selection classifications acknowledged the relationship between plants as descent with modification from common ancestors. Similarities and differences now became evidence of ancestral evolutionary relationships (phylogeny) with god no longer a necessary source of biological order.
Now, in determining phylogenetic relationships, some characters become more important than others. For example, vertebrates are, unsurprisingly, united in their branch of evolution by possessing vertebrae. The production of hypothetical evolutionary trees by computing characters (now often molecular) is known as cladistics, a system now employed throughout biology.
Through the history of the classification of plant taxa certain kinds of characters have been thought of as the key to a fully natural classification – whether they be those of pollen, chemicals etc. It would, however, appear that with the advent of genomics we are currently getting close to the finalization of a final ‘tree of life’ which includes some 350 plant families, and over 350,000 plant species.
From medicine to sustainability
With the ‘tree of life’ near completion, scientific resources once devoted to taxonomy are now addressing other plant questions. The study of plants, across the world, emerged out of the use of plants for medicinal purposes.
From around the 15th century the main task of botany had moved towards a global plant inventory, a task whose foundations had been largely laid by the European colonial powers in the mid-19th century at about the time of publication of ‘On the origin . . . ‘.
There then followed a period of plant science that addressed the remaining questions of plant function through an intense period of laboratory experimentation in plant physiology complemented by advances in field studies as a more scientifically based study of environmental relationships that would become ecology.
By the mid-20th century the mysteries of plant structure and function were, in principle, resolved. The success of the human enterprise, not only in botany, was now becoming evident as an explosion in population with its associated consumption of nature. The world, much of which, at the dawn of the European Age of Discovery only a few generations ago, was wild and unknown, is now mostly converted to human use in an age we now know as the Anthropocene.
Our priority today is the gathering of information that will ensure that environmental damage bequeathed to future generations is minimized.
So, what are the important plant categories, plant classification systems (and their selection criteria) appropriate for the Anthropocene?
The categories we would use for such a classification would relate plants not to one-another, or directly to people, but to human environmental impact. But what would these classification(s) look like?
Commentary & sustainability analysis
We need the most practically efficient way of addressing the future sustainability of life on planet Earth and to achieve this we need to develop a set of priorities. This discussion of plant classification might seem an extraordinarily convoluted way of expressing this but it has given us some insight into the way our minds work and how this has a bearing on both the progress of science and the way we make decisions.
Classification, as the purposive organization of taxonomic units into groups based on prioritised criteria, is an innate characteristic of sentient organisms although consciousness of the process varies in degree. Some classifications are progressive since they can be refined and improved over time, especially those that are shared by a community and can be handed on to future generations. Scientific classifications are progressive and have told us much about the physical world but to address modern problems such as that of sustainability, as a taxonomic goal, is a complex matter involving contentious taxonomic units.
Plants as objects of physical and spiritual significance – as foods, tools, structural materials, and medicines, but associated with this was a spiritual realm that became particularly associated with the effect of plants on the human body. Plant knowledge that went beyond the everyday, plant knowledge that connected to the spiritual world, became the preserve of a special individual, a shaman or medicine-man of some kind.
With the advent of cities much traditional knowledge was lost: knowledge of individual plants, where they grew, and their particular association with the environment and other plants and animals. Cities resulted in a completely new set of categories relating to human-created physical space where plants could be managed (cultivated) – new categories relating to gardens, parks, fields, burial grounds, markets, roads, houses, palaces and so on.
- Classification is the ordering or grouping of objects by criteria that satisfy an end or goal.
- Taxonomy is the study of classification, its principles, and procedures.
- Classification begins with the pre-conscious structuring of our experience that occurs before conscious deliberation. This intuitive ordering of the world is part of the way we represent the world and part of our metaphysical intuition about the world. It is also part of our human reality, our uniquely human collective mental umwelt.
- Biological classification of organisms arranges them by their presumed evolutionary relationships
- One method of scientific advancement is the designation of an are of study (an academic discipline) with a community of scientists that constantly refine the categories, principles and procedures used by that study
- Plant classification is the arrangement of plants into groups according to selection criteria that serve a particular purpose. However, in common usage, the expression ‘plant classification’ refers to the organization of plants into groups that reflect their evolutionary relationships as plant taxa which, prior to Darwin and the theory of evolution, involved the organization of mostly morphological characters as similarities and differences.
- We study plants through the broad range of scientific disciplines that fall under the heading of plant science (formerly the narrower discipline of botany). There is no current and widely accepted system of categories, principles and procedures for the study of the topic ‘plants and people’
category – any unit of experience that can be used for categorization
categorization – the organization of categories
classification – the arrangement of objects into groups by criteria that depend on the purpose of the classification
focus – the restriction of our attention to a limited range of representational units so that, at any given time, our attention is divided into a foreground and background
purpose (taxonomy) – the reason for, or goal, of a classification
nested hierarchy – the taxa and their groupings are contained within other groups in a boxes-within-boxes fashion, like a Russian doll
rank-value – the ranking (valuing or prioritizing) of objects relative to one-another within a classification system.
segregation – the organization of experience into meaningful representational units, both those of cognition (concepts), and those of perception (percepts)
taxonomy – is the study of the principles and procedures of classification
First published on the internet – 1 March 2019
… revised 11 January 2021