Some Linguistic Aspects of the Term “Statistics”

Definition

The origin of the word “Statistics” is mistakenly sought in the Latin word “Status”. Although the term is indeed etymologically linked to “Status”, it is derived from the Greek noun “Στατός” (statos), meaning standing.

1. Introduction

Pawitan and Lee start their recent article on “The philosophy of statistics” with the phrase, «Deep down, we are all interested in truth» [1] (p. 41). Reading this sentence, I wondered what we know about the history of the word “Statistics”? What does it mean? What is its etymological origin?

A word cannot be isolated from its historical background. The need for a new word is scientifically determined, and its history is, partly, a matter of etymology. Through the science of Etymology, the study of the origin of words and how their meanings have evolved, we understand the history, i.e., the origin and evolution of each scientific term. The same happens with the term “Statistics”.

The etymology of statistical terms provides us with valuable insights into the history and development of statistical concepts, highlighting their origins, evolution, and connections to other fields. This understanding enriches our appreciation of Statistics as a discipline and enhances our ability to apply various statistical methods accurately and efficiently.

Referring to historical sources, we can trace the development of the fundamental concept of the term “Statistics”. In all sources, from the most prestigious dictionary of the English language, the Oxford English Dictionary [2], to the most popular ones like Wikipedia [3], and from the oldest [4] to the most recent [5], we will read that the word “Statistics” comes from the new Latin word “statisticum collegium” (“council of state”) and the Italian word “statista” (“politician” or “states-man”), reflecting its origins in state governance and population data collection.

The question that arises is, since “Statistics” is a process of counting, presenting, and analyzing data, why was it connected to the “state” of “state governance” and not to “governance”? “State” is a very general concept, while “governance” is an administrative process. “State” needs “governance” to function. “Governance” is what needs Statistics to govern the “state” effectively.

2. The Origin of the Term “Statistics”

Going through the numerous texts in the statistical literature on the origin of the term “Statistics”, we will realize the following:

i.

In 1633, not 1589 as some scholars erroneously state [6] (p. 447), Gerolamo Ghilini (1589–1668) used the word “Statistics” as an adjective in the title of his book, “Ristretto della civile, politica, statistica e militare scienza”.

ii.

According to the Encyclopedia of Social Sciences, the earliest known occurrence of the word “Statistics” is in the 1660 lectures by Hermann Conring (1601–1681) [7] (p. 356).

iii.

In 1672, a book entitled “Microscopium Statisticum’s full title is Microscopium Statisticum: quo status imperii Romano-Germanici cum primis extraordinarius, ad vivum repræsentatur”, was published in Germany under the pseudonym Helenus Politanus. The author’s identity remains unknown. [5] (p. 6)

iv.

Before 1749, the terms “political arithmetic” and “tabular arithmetic” were used to denote anything that has to do with the present use of statistics.

v.

In 1749, the German professor Gottfried Achenwall (1719–1772) was the first one to use the term “Statistics” in his work entitled “Statsverfassung der heutigen vornehmsten Europaischen Reiche und Volker” [8]. Achenwall used the word “statistik” to refer to the collection and tabulation of numbers about the state.

vi.

In 1770, the term was used by W. Hooper when he translated J. F. von Bielfeld’s (1717–1770) “The Elements of Universal Erudition”. Bielfeld, and hence Hooper, did not mention Achenwall [9].

vii.

In 1787, E. A. W. Zimmermann (1743–1815) introduced into English the term “Statistics” in his “A political survey of the present state of Europe”. Zimmermann, contrary to Bielfeld, mentioned Achenwall and his term “Statistics” [10] (p. 1).

viii.

In 1790, Sir John Sinclair (1754–1835) and his work “Statistical Account of Scotland” established the term “Statistics” as we understand it today [11].

The interested reader is referred to [4,6], among others, for a comprehensive presentation on the origin of the term “Statistics”.

3. The Etymology of the Term “Statistics”

In statistics, precise language is crucial. Etymology, the study of the origin and history of words, aids in developing precise and unambiguous terminology. Knowing the roots of statistical terms can provide deeper insight into the concepts they represent. Accordingly, etymological knowledge can facilitate teaching and learning by making statistical terms more memorable and meaningful. Students who understand the origins of terms may find it easier to remember and apply them.

During the 19th century, many scholars were involved in lengthy discussions concerning the meaning and derivation of the term “Statistics”. It is characteristic that Mohl, quoted by John ([12], p. 656), called it «a psychological curiosity». The vast majority of statisticians were inclined to the opinion that “Statistics” was derived from the Latin “status”. See [12] for a comprehensive presentation of the issue. The jewel of the crown of this dispute was a manuscript that was found in the library of the University of Gottingen where Achenwall had written, «Statistics does not come from the German word Staat but, is a word that is not new. It was already in use in the XVII century and has its origins in the famous Italian “ragion di stato”. The Italians, from “ragion di stato” coined the word “Statista, i.e., Man of the state, Staatsmann”. German language writers adopted this term, and thus the name of this branch of learning… It is mistaken to write Staatistica». The manuscript was headed “An etymological argument”, and it was considered a legacy left by Achenwall [12,13].

Currently, if we ask any colleague of ours, «From where do we get the word “Statistics”?», it is more probable to receive the following answer: «The word “Statistics” is derived from the Latin “Status” used in the sense of a political state». “Statistics” is established as information for the state. Therefore, the relationship Statistics–State–Status, adopted by most of us, is considered given and unquestionable. To endorse this statement, let us see a few representative references in alphabetical order:

• The American Heritage Dictionary of the American Language informs us that the word “Statistics” derives from (a) the German “Statistik”, i.e., political science, (b) the new Latin “Statisticus”, i.e., of state affairs, (c) the Italian “Statista”, i.e., person skilled in statecraft, (d) the old Italian “Stato”, i.e., state, and (e) the Latin “Status”, i.e., position, the form of government.
• Bessant and MacPherson report, «the term statistics … concerns the collection of numerical and non-numerical information about the state (from which the word was originally derived)» [14].
• Lovric states in his International Encyclopedia of Statistical Science in the entry “Statistics: Origin of that Term”, «It is widely believed that the term statistics originated from the Latin Status (situation, condition) of population and economics; in late Latin, the same term meant State. Another root of the term comes from the Italian word stato (state), and a statista (a person who deals with affairs of state)» [15].
• Martynenko writes that, «The point is that the mystifying interpretation of the term statistics correlates to its primary meaning, which arose in the middle of the 18th century for the description of political, economic, demographic, etc. state of a country. “State”, a word of Latin origin, rooted in Germanic languages (cf. German “Staat”), and acquired the meaning “mode or condition of being” at the late stages of Latin. This status was then described by the discipline called political arithmetic and later by state science» [16] (p. 107).
• Miller states the words “Statistic”, “Statistical” and “Statistics” «come indirectly from the mediaeval Latin “status” for a political “state”. More directly statistics entered English from the German Statistik, as a term comparable to mathematics or ethics» [17].
• Finally, in Schwartzman, we read, «The word “statistic” is defined as follows: “statistic (noun), statistics (noun), statistical (adjective), statistician (noun): the Latin verb stare meant the same as its English cognate stand; both are descended from the Indo-European root sta– “to stand.” From the Latin verb came the noun status, literally “a standing, a condition.” The word state, which comes from status, is therefore a synonym of condition. A nation as a whole is also called a state because it is composed of the set of all conditions in a certain geographic area. As a result the term state came to be associated with government and politics…. Although the etymology of the word statistics is complicated, the meaning of the underlying Indo-European root is still apparent: statistics tell you how things “stand”.”» [18] (p. 206).

Schwartzman stopped his investigation one step earlier. Nothing is complicated about the etymology of the word “Statistics”. The so-called, by Schwartzman, Indo-European language is simply the Greek language which has the root sta–, meaning “to stand”. The Latin word “Status” is not original but originated from the Greek word “Στάσις” (pronounced stasis), meaning placement, posture, setting, or stand. From “Στάσις” we get the word “Στατός” (pronounced statos), meaning set, placed, and standing, the word “Στατικός” (pronounced statikos), meaning causing to stand or bringing to a stand-still, and the verb “Στατίζω” (pronounced statizo), meaning to place, to stand, and to stack. All these Greek words come from the Greek verb “Ἵστημι” (pronounced histimi), meaning to stand. A fine statistical example is the term “histogram”, literally a ‘standing up graph’ [19]. All translations from the ancient Greek words to their present-day English equivalents are derived from Liddell and Scott [20], which is considered, by all scholars, as the authority dictionary of the ancient Greek language.

The assembled vertical stacking allows us to easily measure and record the stacked product. Stacking is perhaps the first and simplest statistical procedure/application. Therefore, the term “Statistic” refers to the stack, not the state.

To the best of my knowledge, the first one to point out the Greek origin of the word “Statistics” was Bissell, when he wrote, «The origins of words in current use often hold surprises. Statistical terminology is no exception, … how about “statistics”? The ultimate origin is in the Greek for the science of weighing, one way of assessing the condition of an object. From thence, it moved to the political sphere of ‘status’ (thirteenth century), “estate” of the realm (fourteenth), the territory thereof (sixteenth). A statist was a politician or states-man of the sixteenth century, and the facts concerning the state were termed statistics in the eighteenth century. Other related words have political links—station, statute, and status quo; stationer, stand, statue, statics and stature are also related» [21] (p. 213).

4. From “Setting Up” to “Statistics” via “Stacking”, “Balancing”, “Weighing” and “Measuring”

The ancient verb “Ἵστημι” had an interesting evolution in the Greek language. It originally meant “to set up”, “to place”, or “to make stable”. That is, the basic meaning of the verb was the action of placing something in an upright position. For example, erecting a pillar or a statue.

From the meaning of “set up” came the meaning of arranging or gathering things in one place. Thus, it was also used in the sense of “to stack”, i.e., to place objects in order or in piles. Stacking objects on top of each other can be considered a form of “stacking” multiple objects together.

From the concept of “set up”, the idea of balance evolved. When something is properly set up, it is in balance. This concept of balance was linked to the idea of weighing, that is, placing something on a scale to find the balance of its weight. In ancient times, the scale was a tool that used the concept of setting up to find the balance of weight.

From the idea of weighing, the word extended to measurement in general, as weighing is a form of measurement. Balancing items on scales was critical to weighing. Thus, the word “Ἵστημι” came to be used in the sense of “to weigh” or “to measure”, as to set the scales and balance the weights.

The root sta– is inherent in the term “Statistics”. In the ancient Greek language, we find several composite words whose meaning includes the process of measuring and recording and a second component with the root sta–. Related examples are the words Ovolostasia, Velostasis, Zygostasis, and Psychostasis [20].

The word “Obolostasia” comes from the words “obolos” and “stasis” and usually refers to the process of weighing and verifying the weight of obolos, which were ancient Greek coins. This process was important to ensure the authenticity and value of coins in transactions. Later, the word extended to mean usurer’s profession or usury since the usurers stacked, weighed, and counted obolos. The word “Velostasis” is related to the movement of arrows (velos in Greek) or other missiles, i.e., range of warlike engines or battery of warlike engines. “Zygostasia” is a term derived from the words “zygos” (weighing scales) and “stasis”. It is used to describe the act of weighing or the process of balancing and evaluating.

“Psychostasis” is a very interesting case from a statistical standpoint. Psychostasis or Psychostasia is a composite Greek word meaning the weighing of souls. Its first component is the Greek word “Ψυχή” (psyche), meaning the soul. Psychostasis is a concept that comes from ancient Egyptian and Greek religions and mythologies and refers to the weighing of the soul after death. It is a process where the soul of the deceased is placed on a scale, and the outcome of this weighing determines the fate of the soul in the underworld or the next life. Historically speaking, Psychostasis is the earliest statistical procedure since it incorporates the collection of data, analysis of collected data, presentation of results, and interpretation of results [22].

5. Concluding Discussion

Apparently, “State” and “Statistics” go hand in hand. This is more than obvious if we go through papers referring to teaching statistics. Prime examples are «Statistics has its origin in the word “state”, and originally meant any facts and figures connected with the running of a state» [23] (p. 71) and «The precise origin and meaning of the word statistics is unclear. There are various possibilities. It probably derived from the Latin word status which meant the state of things but could also mean a state. Another possible origin is the Italian word “statera” meaning a balance. All these connotations applied to the discipline founded by Conring. However, another possible origin lies in an Italian tradition of the end of the 16th and the beginning of the 17th century known as “Ragione di Stato”. This encompassed publications concerning state legislation, the distribution of power, its geography and so on. Knowledge of the relevant “Ragione di Stato” was considered essential for the statera or statesman to gain insight into practical politics. Achenwall would certainly have known of this tradition, and may have exploited its name in coining the German term “Statistik”» [24] (p. 10).

To date, the etymological origin of the word “Statistics” is mistakenly sought in the Latin word “Status” (society, state) and refers to the recording and record keeping of quantitative items. The truth is, the term “Statistics” is etymologically linked to “State”, and it is derived from the Greek noun “Στατός” (statos), meaning standing, and/or from the Greek verb “Στατίζω” (statizo) meaning to stack.

As statisticians, we must be precise and realistic. Hence, it would be very unrealistic of me if I would not state that while the etymological origin of the term “Statistics” is perhaps of interest to some readers, I believe that it will only be of very limited interest to most. The term “Statistics” is indeed derived from Greek, but it had no bearing on the evolution of the concepts of probability and statistics. Whether Achenwall and Bielfeld thought they were borrowing from Latin or Greek did not determine the sense they gave to their newly coined word.

Given my above argument, is this paper justified to be written? As both an answer to this question and an epilogue, I quote Cuoco and Rotman’s explanation for the usefulness of etymology in all mathematical sciences. «Etymology … traces out the origin of … terms. We believe that knowing the etymology of terms often helps to understand the ideas they name» [25] (p. xiv).