Per Capitalization

Individual words do not usually trigger the same devotion as, say, favorite songs. Yet some people find the time to have a favorite word. A TV host conducts interviews famous for including the question, “What is your favorite word?” His guests know it’s coming, so they are prepared with a word or phrase chosen for either meaning or sound. I have wondered if the favorites are situational. Who can have just one favorite word?

The “sound” choices indicate an appreciation for the particulars of articulation. That is where my instinct points too. No one has asked me lately if I have a favorite word. Then again, I have not been interviewed on Inside the Actor’s Studio. However, there is one phrase I have had to think about so much that it has become a favorite.

“Per capita” occupies a plot of linguistic real estate in my head. Not only is “per capita” a trifecta of plosives (bilabial, lingua-alveolar, and lingua-velar) separating the vowels with a postalveolar “r” thrown in, the phrase also has a pleasing rhythm. It is fun to say and snaps with the summery crispness of the waffle cone holding your scoop of pralines and cream.

Best of all, it presents a hotbed of style issues just waiting for AMA Manual of Style application.

“Per capita” became the contested part of an axis label earlier this year. While not quite an axis of evil, the label was rather vexing. All roads lead to Rome, and all questions lead to the AMA Manual of Style. In Section 4.2.7, “Axis labels in figures are akin to column headings in table, so each word should be capitalized (except minor words such as prepositions of less than 4 letters).” As first presented on the axis label, “per” followed style for a minor word and was thus lowercase. One reviewer suggested initial cap “Per,” which met with the author’s enthusiastic approval. After all, “per capita” is not only a minor preposition with a major noun; it’s so much more! The manual refers to Merriam-Webster’s Collegiate Dictionary as an authority. “Per capita” receives its own dictionary entry, so AMA style considers it a single entity. I went along for the ride, on the right track to make an author happy.

A round later, another reviewer indicated that “per” is never initial cap in JAMA Network journals. I searched “per capita” on The JAMA Network site and the hits from 2010 to the present indicated roughly half with lowercase “per,” roughly half with uppercase “per,” and an outlier with initial cap and hyphen. It appeared that style was up for grabs. I reverted to lowercase, asked for the author’s understanding, and moved on.

Not long after, I mentioned the exchange to an AMA veteran. After a pause, he said, “Well, maybe per capita is considered a foreign phrase like in vitro.”

Wish I had thought of that.

I had lost sight of the origin: Latin for “by the head,” an obvious parallel to Latin for “in glass.” The pesky “per” had sidetracked me. In Section 10.2.1, “In compound terms from languages other than English, capitalize all parts of the expression.” Hence Per Capita. Of course Per Capita reappeared, the next  time in the title of a reference to a book. Now I will be ever alert to “per capita” wherever it may need to be in title case.

The takeaway is that “per capita” in an axis label illustrates several manual of style tenets: follow the latest edition of Merriam-Webster’s, consider word function (stand-alone “per” which is lowercase vs compound term “Per Capita”) in applying the “minor words” rule, and capitalize all parts of a compound term from another language. Not a bad litany of accomplishments for 4 syllables.

Back to favorite words. An article on that topic might not easily fit into one of the network journals, but I would like to see a figure that plots the number of favorite words for each capita in a study population. (The style guide does not rule on whether “capita” can be sold separately from “per” but “for each” seems a good stand-in.). When the day comes that a JAMA Psychiatry article presents a figure depicting Number of Favorite Words Per Capita in Adults Who Edit, I will know this journey has come full circle.—Timothy Gray


“Recognizing that the genetic contribution to health disparities is likely to be relatively limited is not the only reason to question the wisdom of promoting genetic research under the aegis of health disparities.”1

“In several examples published under the aegis of the [Agency for Healthcare Research and Quality], the benefit side rests on the data from randomized controlled trials….”2

“[T]his definition and associated regulations have become de facto rules for US academic and other research institutions and are applied to any work done by their employees or under their aegis regardless of the source of funding.”3

While somewhat infrequently encountered in medical writing, aegis is occasionally used in content related to health policy, institutional oversight, or the conduct of research. It also is occasionally used to refer to a journal’s vouching for the validity of data or other findings published in its pages; when such validity is called into question, a journal may issue a full retraction or simply “withdraw aegis,” ie, issue a statement to the effect that the journal is no longer responsible for the data.4

In its original sense, aegis (Greek aigis [goatskin]) referred to the shield or protective cloak worn by Zeus or Athena in the myths of ancient Greece.5 In the centuries since, the word has by association come to be used idiomatically to indicate defense (“Feeling is the aegis of enthusiasts and fools.”6) or guidance or influence (“They made their valuable individual contributions, but under the Ellington aegis they found themselves constantly enriched musically.”6). Similarly, the word is perhaps most often used in the construction “under the aegis of” to express patronage or sponsorship (“under the aegis of the museum”) and, especially, protection (“a child whose welfare is now under the aegis of the courts”).5

All of which is perfectly comforting. However, the classical roots of the word are far less benign. Whereas aegis-like shields appear in Egyptian, Nubian, and Norse mythology and art,7,8 the most elaborate account of the origins and attributes of the aegis comes, as suggested above, from the narratives of ancient Greek mythology and literature, in which it is consistently depicted as an object possessing fearsome supernatural power.

In one such narrative, the goat deity Amaltheia suckles the infant Zeus, who then—in the curiously detached manner so frequently encountered in mythological accounts—breaks off her horns and flays her hide. From one of her horns Zeus fashions the cornucopia, or horn of plenty; from her hide, he fashions the shield or cloak that would come to be called the aegis, which he then wears, on the counsel of an oracle, into battle against the marauding Titans.9 However, in the hands of Zeus, king of the gods, the aegis is more than a protective device; when in his wrath he shakes the aegis from his perch atop Mount Olympus, thunder crashes, bolts of lightning slash the sky, and fierce storms devastate the land.10 Moreover, in the Iliad, Homer describes the device as the “tempestuous terrible aegis, shaggy [and] conspicuous… given to Zeus to the terror of mortals.”11

But the aegis is associated with deities other than Zeus. In the Iliad, for example, Zeus lends the aegis to Apollo, who wields it to push his enemies back to their ships12; in other accounts, Zeus lends the aegis to his daughter, Athena, goddess of war,12 or presents it to Athena after his conquest of the Titans.9 In the most colorful account, Zeus swallows his wife, Metis, whole—after which Athena is born from Zeus’ head, emerging fully formed and bearing the aegis and other weapons of war.13 Regardless of how Athena comes to possess the aegis, however, in her hands it becomes more formidable still. For example, in the Iliad Homer reverently describes the aegis as no mere goatskin but rather as “ageless and immortal,” worn by “bright-eyed Athene” and adorned with tassels of the purest gold12; moreover, elsewhere in the Iliad he describes the aegis as a dramatic golden cloak fashioned by Hephaistos, god of fire and metalwork,14 to resemble a scaly skin like that of a snake, linked and fringed with writhing serpents, and bearing in its center the severed head of the Gorgon Medusa, eyes rolling and scalp also bristling with serpents.11

Charming as this may be, given the provenance of the word and the fell associations that come with it, how did aegis, apart from its military associations, come to be used to express defense or protection? Perhaps more puzzling, how did it come to be used to express benevolent guidance, influence, or patronage? A possible clue is that whereas early accounts often depict Athena as cultivated, civilized, urbane, and wise, the economy of ancient Greece was bolstered by military pursuits, and in short order Athena came to be depicted as the goddess of war—although, importantly, her military might was tempered by the cultivation and divine wisdom earlier attributed to her.15 Thus, Athena is sometimes viewed as having 2 sides. The first is characterized as wrathful, tempestuous, and destructive; however, the other is characterized as divinely beneficent and endowed with the inclination and ability to grant the gifts of aid, wisdom, and protection to favored mortals.16 The power of both sides is wielded through the supernatural power of the aegis—hence Homer’s description of the device in the Iliad and its depiction in visual art as inky black or brilliant gold.11,12,16

Interestingly, however, while Athena extends divine beneficence and protection—backed by the wrathfulness also attributed to her—she does not authorize those mortals fortunate enough to come under her favor to act on her behalf. Thus, whereas in current usage aegis is correctly used to express defense, protection, guidance, influence, patronage, or sponsorship, its use to indicate “under the jurisdiction of” is considered incorrect.17Phil Sefton, ELS

1. Sankar P, Cho MK, Condit CM, et al. Genetic research and health disparities. JAMA. 2004;291(24):2985-2989.

2. Vandenbroucke JP, Psaty BM. Benefits and risks of drug treatments: how to combine the best evidence on benefits with the best data about adverse effects. JAMA. 2008;300(2):2417-2419.

3. Iverson C, Christiansen S, Flanagin A, et al. AMA Manual of Style: A Guide for Authors and Editors. 10th ed. New York, NY: Oxford University Press; 2007:157.

4. Hammerschmidt DE, Franklin M. The limits and power of peer review. Minn Med. June 2006. Accessed February 13, 2013.

5. Aegis. Merriam-Webster’s Collegiate Dictionary. 11th ed. Springfield, MA: Merriam-Webster Inc; 2003:19.

6. Aegis. The Compact Oxford English Dictionary. 2nd ed. Oxford, England: Oxford University Press; 1991:22.

7. Aegis: in Egyptian and Nubian mythology. Museum of Learning Web site. Accessed February 15, 2011. 8. Aegis: in Norse mythology. Museum of Learning Web site. Accessed February 15, 2011.

9. Amaltheia. Theoi Project Web site. Accessed February 13, 2013.

10. Zeus. Theoi Project website. Accessed February 13, 2013.

11. Hephaistos Works 2. Theoi Greek Mythology Web site. Accessed February 13, 2013.

12. Teleporter: aegis mound. Entropia Planets Web site. Accessed February 13, 2013.

13. Athena. Theoi Project Web site. Accessed February 13, 2013.

14. Hephaistos. Theoi Project Web site. Accessed February 13, 2013.

15. Athena. Encyclopedia Britannica Online. Accessed February 13, 2013.

16. Deacy S, Villing A. What was the colour of Athena’s aegis? J Hellenic Stud. 2009;129:111-129. Accessed February 13, 2013.

17. Bernstein TM. The Careful Writer: A Modern Guide to English Usage. New York, NY: Athaneum; 1985:29.

Questions From Users of the Manual

Q: What is the difference between “percent” and “percentage”?

A: We regard “percent” as being a unit (equivalent to “kilograms”) and “percentage” as being a description of something that is measured in percent (equivalent to “weight”). As a rule of thumb, this would mean that “percent” (or the percent sign) would usually be used after a number: “In 10% of participants…” When discussing these values more broadly, “percentage” would be appropriate: “The percentages in Table 4 are from the study results in 2007.”

Q: When a percentage is associated with a drug, does it belong before the drug name or after it? The examples used in sections 15.4.9 and 15.4.10 seem to contradict each other.

A: Typically, the percentage would follow the drug name, as described in section 15.4.10: metronidazole lotion, 0.75%. The example in section 15.4.9 is slightly different as it describes the percentages of the components in a single product: “an artificial tear product containing 0.42% hydroxyethylcellulose and 1.67% povidone.”

Q: The 10th edition refers to both “press release” and “news release.” Are these used interchangeably or is there a distinction?

A: We consider these interchangeable terms, but in the next edition we may use “news release” exclusively because information is distributed well beyond print media.

Q: Do you drop the periods in the abbreviation LLC (limited liability company)? I see that you recommend dropping the period after Co (company) and Inc (incorporated) and wondered if this would be treated similarly.

A: Yes.

Q: Your manual (p 342) recommends that a colon not be used after because or forms of the verb include. Does this recommendation include situations in which the word include precedes a bulleted list?

A: Yes, it would be applicable whether the copy that follows the verb include precedes a bulleted list or run-in text. The key is that a verb should not be separated from its object or predicate nominative. For example, we would recommend the following:

There are many treatments for skin irritation. The treatment prescribed may include creams, sprays, and gels.

There are many treatments for skin irritation. The treatment prescribed may include

  • creams
  • sprays
  • gels

Another option would be the following:

There are many treatments for skin irritation: creams, sprays, gels.

—Cheryl Iverson, MA

Let’s Not Be Rash

“I have a rash all over my back.”

“Physical examination revealed a diffuse rash covering most of the patient’s back.”

One sentence overheard in conversation, the other appearing in a medical report; the former more colloquial, the latter more technical—but both correct, right? Well, not quite. While rash has been used in medical writings since at least the second half of the 19th century, current thought is that rash is a lay term lacking precision and specificity. Thus, while it is perfectly acceptable (and understandable) for a patient to use the term when describing symptoms to a clinician, once the clinician turns to describing an inflammatory (as opposed to a cystic or neoplastic) skin condition for the patient’s record or in a case report or other formal medical literature, rash should be replaced with a more descriptive and specific term.

However, this does not mean that rash has no place in medical literature. Because patients typically use rash when describing their symptoms, the term might legitimately appear, for example, in a report of a patient presenting with specific symptoms and signs or a list of adverse reactions self-reported by participants in a randomized trial; similarly, report forms provided to trial participants might also include rash in a list of potential adverse reactions.

Editors of medical literature, then, need to take into account the context in which rash was originally used as well as the intended audience. The term is appropriate when it appears in passages quoted from patients or when used in materials intended for lay use, eg, instruction sheets, patient care brochures, and medication inserts (however, use of the term skin rash in such materials is redundant and should be avoided [see 11.2.1, Redundant Words, in the AMA Manual of Style, pp 405-406 in print]). In many cases, as noted above, the term is also appropriate in more formal medical literature reporting nondermatological studies in which rash is one of several adverse events listed and any type of rash would be lumped together in one category. In literature describing physical examinations, specific skin-related adverse events, or dermatological studies, rash should be replaced with a more precise term. While skin eruption or dermatitis are often acceptable replacements, ideally the replacement term should be precise enough to describe the appearance of the lesion (eg, macule or patch vs papule, bulla, vesicle, nodule, or pustule) and the suspected specific cause (eg, allergy; heredity; irritation; bacterial, viral, and fungal infections). Replacement terms might also describe the distribution of inflammation and the histological structures involved.

The range of findings and topics reported in JAMA and the Archives Journals allows for the use of rash as well as more specific terms within a single family of publications. For example, rash does appear regularly in JAMA, since many studies in that journal report adverse events that include rash as a category. On the other hand, the editorial staff of the Archives of Dermatology always replaces rash with a more specific term, unless rash appears in material quoted from a patient.— Phil Sefton, ELS