For Whom and by Whom Children Are Named: Family Involvement in Contemporary Japanese Naming Practices

Abstract

In pre-modern Japanese naming practices, familial relationships were frequently demonstrated systematically through personal names, but with changing lifestyles, family structures and naming trends, such systematic ways of creating familial ties through personal names have largely been lost. However, personal names may still express familial ties, but in different ways than in previous times. To consider this, this article utilizes a unique data set of 303 messages in municipal newsletters from parents about how they chose their child’s name, focusing on who was listed as choosing the name; whom the child was named for; and common elements within parent–child pairs and sibling sets. Parents themselves were most frequently noted to have selected the name, followed by the child’s older siblings; in comparison, grandparents were listed rarely. The use of a shared kanji ‘Chinese character’ between parents and children was also not common; however, it was more frequently observable in siblings’ names. Although the data set is small in size, the data strongly suggests that contemporary families are focused more on creating intragenerational connections between siblings, rather than intergenerational familial ties, which may be a result of the nuclearization of contemporary families.

1. Introduction

Although we may feel like our names are an inseparable part of ourselves, names are neither natural nor inherent, in so far as they must always be given, selected or chosen. Names create a web of relationships, both through the act of naming itself and through the choice of names. As Benson (2009, p. 180) notes, our names speak more of the people who gave them to us than they do about ourselves, as those giving names choose names to recognize other people or infuse names with meanings personal to themselves. Because naming practices embed children within societies (Bodenhorn and vom Bruck 2009, p. 3), observing naming practices and trends can help us examine what people desire for and expect of children (S. Tanaka 2014, p. 5). When naming practices inevitably change over time, these changes not only reflect the social systems and institutions of their times, but also how those social systems and institutions themselves have changed (Plutschow 1995, p. 1). As familial relationships are also social systems, changes in naming practices can also reflect changes in the relationships between those who are involved in naming.

This point is particularly relevant for naming in Japan as contemporary naming practices have changed tremendously over the past few decades. New types of names have appeared that have received highly negative public reception (overviewed in (Unser-Schutz 2016a)), and previously popular forms such as female names ending in –ko, which used to be viewed as the typical female name, are now not commonly given (Hashimoto and Itō 2011; D. Kobayashi 2001; Komori 2002). These changes in naming practices have led to a generational divide in names. The types of names commonly found amongst young children are now very different from those of their parents; likewise, the names found amongst the parents of young children and their own parents—that is, the children’s own grandparents—are also very different, meaning all three generations have very different types of names (see Unser-Schutz 2016a) for an overview of naming practices in the 20th century).

Given that names are generally selected by parents, the changes observable amongst young children’s names speak to a gap between the naming preferences of young parents and previous generations. As such, it is not surprising that changes in naming practices have been attributed to changes in values, familial relationships and family structure. Recent names overlap with an increased prioritization of uniqueness and individuality in contemporary Japan (Ogihara et al. 2015; Y. Kobayashi 2009), while deep decreases in names indicating birth order (Honda 2005), such as Ichirō ‘first-boy’ and Saburō ‘third-boy’ are likely due to smaller family sizes (Nishino 2009). Linking these two points, it has also been suggested that families are giving increasingly unique names because, with smaller numbers of children per family, each occasion of naming is viewed as being more important (Tahara 2008). It has also been suggested that such new names are the result of changes in who is involved in naming, with grandparents in particular less actively involved in the naming of children (Otake 2012).

To evaluate how such changes in familial relationships are reflected in changing naming practices, this study uses naming data from municipal newsletters of messages on how names were chosen for local children to observe who is involved in the naming process, who children are being named for, and commonalities between children’s names with their parents and siblings. As a result of the analysis, it will be shown that naming primarily involves the nuclear family of parents, children and their own siblings, with grandparents and other family not actively reported to being involved in the process. Similarly, children were not frequently named after people not in the nuclear family. Compared with the commonalities between children’s names and parents’ names, commonalities are more regularly observable in the names of children and their siblings, which can be read as a way of strengthening ties within generations, as opposed to between generations. This may support arguments that naming trends are changing partially because older generations are not involved in naming. This may suggest that the values associated with previous naming practices are not being inherited; at the same time, today’s naming practices may result in the strengthening of ties within the nuclear family. Note that this article is based in part on Japanese-language working papers, primarily Unser-Schutz (2017b), with an advanced data set and analysis.

2. Background on Japanese Naming Practices: What Is Prioritized When Giving Names?

One of the major characteristics of Japanese naming is the fact that it is very open to the creation of new names, especially in comparison to languages such as English (Honda 2005). Modern naming practices formed in the period of Westernization in the Meiji era (1868–1912), when two major laws were passed in 1872 (1) limiting individuals to one family name and one personal name, and (2) making it illegal to change names—a practice which had been common as people (or rather generally, men) went through major life events such as becoming an adult (see Ōtō 2012) and (Plutschow 1995) on historical naming practices). Aside from these limitations, however, Japanese parents are given great freedom in the selection of children’s names, especially when compared with, for example, Iceland, where one must be able to prove that the chosen name is previously extant and matches the sex of the child (Willson 2009), or Tajikstan, where individuals are obliged to choose from a list of 3000 possible names (Asia-Plus 2015).

The only major legal limitation regarding choice of names concerns what orthographic scripts may be used. Although Japanese is written with a combination of (1) kanji ‘Chinese characters (ideograms)’, (2) the katakana and hiragana syllabaries, referred to collectively as kana, (3) romaji ‘the Roman alphabet’, and (4) Arabic numerals, only kanji and kana are permitted to be used in names. The set of kanji legally useable in names is limited to the 2136 jōyō-kanji ‘kanji for everyday use’ and the 861 jinmeiyō-kanji ‘kanji for personal names’ (see Emmanji 2005) on these kanji sets and restrictions); even with such restrictions, the set of all possible kanji and kana still comes to over 9.5 million possible two-letter names. Of course, this is somewhat of an exaggeration. As naming is a social practice, not all of the combinations would be deemed socially acceptable; for example, the Akuma-chan jiken ‘Miss Devil case’ provided legal precedent for not approving names with potentially negative ramifications for children (see Yasuoka 2011). Aside from such ambiguous limitations, given that there are (1) no restrictions on the phonetic aspects of names and (2) kanji can be used to write creative, non-established readings called ateji (see Ariga 1989) and (Wilkerson and Wilkerson 2000) on such creative practices), parents can and do give highly creative names to children.

Nonetheless, as K. Tanaka (1996, p. 82) points out, people experience strong social pressure from their community to give names within accepted bounds. This is observable in the rankings of Japanese popular names for the majority of the 20th century. Using data published by the Meiji Yasuda Life Insurance Company listing the most popular names amongst enrollees for each year since 1912, Unser-Schutz (2016b) showed that by the mid-20th century—following late 19th century changes to naming laws—a kind of standardization of names occurred whereby the majority of popular names each year appeared similar to each other in form and structure. The most obvious example of this was -ko names for women. Although prior to the late 19th century, -ko names were only given to members of the royalty, by 1920 between 80% and 90% of women born had -ko names (Komori 2002). Similar patterns in the number of characters used in names and other structural characteristics can also be found in men’s names (Komori 2002; Unser-Schutz 2016b). The standardization of names is also a reflection of conscious attention to the democratization—in terms of its accessibility to ordinary people, given the burden of learning kanji—of Japanese orthography in the 20th century, which led to restrictions in the number of kanji permitted to be used in names (see Emmanji 2005) on the establishment of the set of kanji for names).

In contrast to this period of standardization, since the last decade of the 20th century the speed of change in the most popular names given each year is accelerating, with fewer names repeating from year to year (Hirayama 2011; Unser-Schutz 2016b). New types of creative names have become increasingly popular. However, these new types of names have been subject to intense criticism, particularly because many app2017ear to use kanji in ways that deviate from standard usage, making them difficult to read (Satō 2007; Tokuda 2004; Unser-Schutz 2017a). Examples of this are given in Figure 1, taken from the larger study on which this article reports. Usually, kanji are associated with several different readings, and the Japanese Ministry of Education has provided lists for the appropriate readings for each kanji to be studied at school. In each case in Figure 1, however, the readings of the names—here, Hibiki, Moka, and Reon—cannot be obtained by combining any of those readings to the kanji used. As a result, there is uncertainty in how they are to be read, leading some to suggest that they do not function socially (Satō 2007); some have suggested that they could have a potentially damaging effect on children when they are put in situations that draw attention to their name, such as at job interviews (Makino 2012). As can be seen by the words coined to call such new names—kirakira nēmu ‘sparkly names’ and DQN nēmu ‘ill-educated/stupid names’—the criticism has been extremely harsh, and it has been argued (Unser-Schutz 2016a) that the response to such names can be read as a youth problem, a common in discourse in contemporary Japan (Toivonen and Imoto 2011).

To understand what it means to say that a name might not function socially requires understanding what the function of a name is. The most characteristic function is to identify individuals, but as Jugaku (1990) notes, names can also feature information such as an individual’s nationality; gender; age; region; or—most relevant to the current article—familial relationships. As the phrase na wa tai o arawasu ‘names express their owner’s nature’ suggests, in Japan there has long been a belief that names have kotodama, a classical concept referring to the (mystical) power associated with words (Plutschow 1995). The idea that names have kotodama was reflected in the pre-Meiji period practice of avoiding the use of an individual’s real personal names or imina ‘taboo names’, but a similar belief in the power of names can be observed in contemporary Japan, as parents often choose names that reflect how they want their children to be—as well as Makino’s (2012) assignment of new names as abunai ‘dangerous’.

An additional function of names is to deepen ties with others. Names are in a sense the first gift a parent gives to their child, but it could also be described as the first story parents tell them. According to a survey on how people felt about their names and the origins of their names conducted by Marsh Research (2015), 72% of all respondents (n = 344, total n = 480) knew why they were given their names, meaning that the majority of people have experienced talking about how their names were chosen. It is easy to image that how parents settled upon their child’s name becomes a narrative with which they can tell their children about how much they were anticipating and waiting for their child’s birth. The characters of that story are not only parents and their children, however. Although by law only parents or their representatives can submit birth certificates, parents do confer and receive advice from others, making the act of choosing a name an opportunity for family members and other important individuals to be interested, involved and invested in the child’s life.

No matter how the names are chosen—by influence of popular trends, or through fortune-telling (see Y. Kobayashi 2008)—names always reflect the feelings of the people around them (S. Tanaka 2014, p. 150). In particular, involving older generations such as grandparents-to-be in the naming process can potentially lead to the inheritance of knowledge and naming values. Prior to the Meiji period, it was common for grandparents and other family members to select children’s names—in addition to other practices that involved people beyond parents across Japan, such as having babies themselves select a name from a set of candidates or having priests or wet nurses select names (S. Tanaka 2014, pp. 155–58). Conversely, not involving them can create a gap between generations, and it has been pointed out that intergenerational exchange—particularly between the elderly and children—is becoming rarer, accompanied by an increase in ageism (Thang 2003). The nuclear family has increasingly become the core familial form, starting in the post-war period (see Nonoyama 2000) on family sociology research in Japan; see also Martin 1990 on important limitations when examining demographic household data for Japan), and it has been suggested that this has led to grandparents not being involved in the naming process, thus licensing changes in the types of names parents choose (Otake 2012).

Names themselves also signify relationships between children, family members and other individuals. One way to do this in Japanese is through the sharing of a kanji character. Prior to the Meiji period, it was common in (higher class) clans for male members to share a common kanji character; kanji given to all male clan members were specifically called tōriji, and kanji given to all male clan members of one generation called keiji. It was also common for adult (males) to adopt a family member’s personal name upon inheritance, a practice called shūmei (Plutschow 1995; S. Tanaka 2014). However, with the nuclearization of the family, decreases in the number of children born per family, and the decline of the ie-seido ‘family system’ organizing relationships between families in the same clan (see Hidaka 2011) on the ie-seido), tōriji, keji and shūmei are no longer common practice. The 1872 laws briefed above, as well as the 1874 elimination of the practice of celebrating genpuku ‘coming of age’ for boys between 12 and 16 (Plutschow 1995)—which was generally when imina names including tōriji and keiji would be selected—played no small part in the demise of these practices.

Yet even if such large-scale social practices no longer exist, parents can still form relationships through names by the selection of particular kanji from one parents’ name or choosing some other matching characteristic (the same structure, similar sounds, etc.). Names position a child as a member of their family, but names which are seen as being different may have the opposite effect. When children’s names, their parents’ names, and their grandparents’ names are all dissimilar, this may result in the foregrounding of differences in values and a sense of generational divide, which may contribute to the sense of crisis surrounding new types of names. Given that, as Goodman (2011, p. 164) notes, Japanese youth are Japan’s most important ‘natural resource,’ having young people become integrated into the social system is a crucial issue. Looking at (1) who is involved in the selection of contemporary names and (2) what connections are created with other family members through names can give insight into how family relationships are changing and the social impact of those changes.

3. Methodology

This study utilizes a set of messages from parents in a kōhōshi ‘local municipal newsletter.’ The majority of Japan’s 1788 municipalities produce public newsletters distributed free of charge to residents. These newsletters—which often are high-quality, without much difference in look and feel from other professional printed news magazines—generally include information about local events, issues and other public service announcements (PSA), but they also have additional functions as well. In recent years, there has been increasing demand for municipal newsletters to shift their primary function from PSAs to something more aligned with the community and residents’ needs and interests (Masse Osaka 2013). One type of content frequently seen in municipal newsletters that has a function other than pure PSA is childbirth announcements and related columns introducing children to the community, called here child information columns for simplicity. Child information columns often appear with names such as Wagaya no aidoru ‘Our family’s idol’ (Ina Town in Saitama Prefecture) or Man’issai no goaisatsu ~ wagaya no takaramono ‘Greetings from the under ones: Our family’s treasure’ (Tosa City in Kochi Prefecture), and they generally take the form of a letter from the parent(s) along with a photo and information about the child, including their personal and family names; the area they live in town; their birthdate; and their parents’ names.

As argued elsewhere (Satō 2007; Unser-Schutz 2018), the fact that these newsletters are created and published by the municipalities themselves, which are the authority in charge of the registration of names, means that they could be viewed as one kind of primary resource. In many cases, these newsletters—including back issues—are available online for downloading, making them easy to collect. The majority of child information columns also include glosses on how the names are read, a point previously noted by Satō (2007). In a study of the 1020 municipal newsletters listed at the time of data collection (2014) on the general municipality information site jichitai.com, 50.39% included a child information column, of which 97.28% featured the child’s name in kanji and the reading of the name (Unser-Schutz 2018). Amongst these 1020 municipal newsletters, just one featured a data point of relevance here, which has motivated its selection in this study: As part of its Wagaya no aidoru column, Otobe Town in Hokkaido Prefecture also included messages from parents on why they chose the name, a modified example of which is shown in Figure 2. Given the difficulty of obtaining such detail for any given community over a continued period of time, this is a crucial resource for looking at how names are selected.

A small town, as of the end of January 2019, Otobe consisted of 3716 registered residents (male: 1711, female: 2005) over 1885 households (Otobe Townhall 2019), or an average of 1.97 people per household. Population loss is an important issue: Compared with five years prior in October 2011, the population of 2016 represented an 11.34% (4372) decrease in population. According to data from the 2010 census, of the 4408 residents from 2010, children aged 0 to 14 years old accounted for 11.22% (495) of the population; in contrast, elderly residents over 65 accounted for 34.39% (1516) of the total population, giving a youth population index—calculated by dividing the number of people under 15 by the number of people between 15 and 64 times 100—of 20.7 and an elderly population index—calculated by dividing the number of people over 65 by the number of people between 15 and 64 times 100—of 63.2 (Otobe General Coordination Department, Planning Section 2012). As can be observed by the fact that the elderly population index of Hokkaido on the whole is 39.0, the situation in Otobe is critical. That the Wagaya no aidoru column in the Otobe municipal newsletter—which had previously been in every monthly issue—switched to bimonthly from April 2015 reflects these issues.

Nuclear families are also becoming the norm in Otobe: According to the most recent data from the 2005 census, of the 1898 households then in Otobe, 419 included a child under the age of 18, of which 64 (15.28%) also had a grandparent living with them (Otobe General Coordination Department, Planning Section 2010a). Although Otobe has not explicitly noted why they include a child information column, some insight may be gained from Kyotango City in Kyoto Prefecture. When they started their own column, they specifically wrote that sharing the moving joy at the birth of life and the importance of life itself to residents was part of its plans to develop the city as a place to raise children (Kyotango City Secretarial and PR Department ‘Kyotango News: The Bonds of Life’ Section 2010, p. 29). It is easy to imagine that Otobe had a similar goal in publishing their own column, and the decision to switch to bimonthly issues was not easy, the editors themselves noting that kōhōshi demo ninki kōnā deshita node zannen desu ‘[it] is unfortunate given that [this] was one of the popular columns in the newsletter’ (Otobe General Coordination Department, Planning Section 2015a).

For this study all of the children’s information columns were extracted from an approximately 14.5 year period from April 2004 to November 2018. Naming data was collected for 309 children (F = 155, M = 148), with a total of 303 unique messages after excluding six repeats from siblings sharing one message (one set of triplets, four sets of twins), which included the same messages for (1) in the naming process; (2), an analysis of similarities between matched children’s and parents’ names; and (3), an analysis of similarities between the names of siblings.

Of the two types of messages in the column—the origins of the names and the parents’ desires for children (see Figure 2)—only the earlier were included in this study. Each message was tagged by whether or not they specified an individual who gave the name, and if so, who (e.g., ‘Grandma chose the child’s name’). Each message was additionally encoded by whether or not they specified an individual for whom the child was named; given that being named for another individual in Japanese often involves sharing a common kanji or similar rhythms rather than exactly the same personal name, this was taken broadly to include any time a connection with another individual was noted (e.g., ‘We chose a name that sounded similar to their older sister’s name’). It also includes choices which were inspired by and created connections with others. Many letters noted that they selected the name because a family member liked it, making it ambiguous as to whether to treat the name as being chosen for or by an individual; for consistency, such cases were treated as their family member having been involved in choosing the name, as it indicated that family member’s tastes and preferences were prioritized in the naming selection.

Children’s names were also compared and analyzed with respect to their parents’ names for any similarities in structure, specifically (1) the whole personal name; (2) the number of kanji in their names (such as the son’s name大和 and the father’s name 智明, which are both comprised of two kanji characters); and (3) the use of a common kanji in children’s and parents’ names (such as the character 翔 in the boy’s name 咲翔 and the boy’s father’s name翔太). Comparisons were not made between the phonetic forms of children’s and parents’ names as the phonetic forms were not usually listed in the municipal newsletters.

Finally, siblings within the data set were isolated by matching each child against several data points (the same last names; the same parents’ names; and the same area), which resulted in 85 sibling sets. In addition to the three points noted above for children’s and parents’ names, siblings names were checked for (4) whether they had the same number of morae, a phonological unit that contributes to syllable weight and that is an important element in the creation of rhythm in Japanese that differs slightly from syllabic rhythm (e.g., the names Hotaka, Natsui and Anna are three, three and two syllables, respectively, but are all three morae (Ho-ta-ka, Na-tsu-i, A-n-na); see Warner and Arai 2001); and (5) whether both names included any phonetic similarities, specifically, whether they included the same phoneme, either at the left side (start) of both siblings’ names (e.g., r in Ryūnosuke and Ririko) or at the right side (end) of both siblings’ names (e.g., ne in Kokone and Yuine). Note that for both the comparison between children’s and parents’ names and siblings’ names, all 309 children’s names (F = 156, M = 153) were included.

4. Results

4.1. Who Was Involved in the Naming Process, and Who Were Children Named for?

Of 303 messages, the majority (190; 62.71%) did not list a specific person as being involved in the naming process (

Table 1. Who was noted as being involved in naming the child and their relationship with the child.

Relationship with Child	Child’s Gender 1
	Female		Male		Total
Father	33	21.29%	38	25.68%	71	23.43%
Mother	26	16.77%	18	12.16%	44	14.52%
Parents	12	7.74%	9	6.08%	21	6.93%
Older sister	6	3.87%	1	0.68%	7	2.31%
Family	2	1.29%	3	2.03%	5	1.65%
Older brother	1	0.65%	3	2.03%	4	1.32%
Grandfather	3	1.94%	0	0.00%	3	0.99%
Grandmother	1	0.65%	0	0.00%	1	0.33%
Not listed	94	60.65%	96	64.86%	190	62.71%
Total	155	100.00%	148	100.00%	303	100.00%

¹ Data indicates the total number of messages in which each individual was listed; some messages may list multiple individuals. Listed in order of total percentages.

). Within the 155 messages that did list such a person, an average of 1.43 people (SD = 0.59) were listed. Messages specifying either the child’s father or mother accounted for 71 and 44 messages, respectively, with an additional 21 listing both parents, making the parents themselves the most commonly noted individuals involved in the naming process. Given that the letters themselves were from parents, it seems natural to conclude that this is actually much higher. The second most frequently listed group were siblings, with the children’s own older sisters (7) or brothers (4) listed a total of 11 times (3.63%). Older siblings were actually listed more frequently (as in Example 1) than grandparents, with grandfathers only specified three times, and grandmothers once (as in Example 2).

(1). 

Example 1.Kyōdai to onaji ‘真’ o tsukatte, namae no hibiki to jikaku o mite kimemashita. [[We] used the same [kanji] ‘真’ as [his] siblings, and looked at the sound and number of strokes and decided.] (Otobe General Coordination Department, Planning Section 2011).

(2). 

Example 2.Obāchan ga ‘hana’ ga suki de, yobiyasuku, kawaii namae da to ryōshin de kimemashita. [Grandma likes ‘hana’ [flowers], and [we] the parents together chose it because it was an easy to say and cute name.] (Otobe General Coordination Department, Planning Section 2007).

At 23.76%, messages specifically noting that the child was named after another individual were also a minority (

Table 2. Who the child was named for and their relationship with the child.

Relationship with Child	Child’s Gender 1
	Female		Male		Total
Older sister	23	14.84%	4	2.70%	27	8.91%
Older brother	6	3.87%	14	9.46%	20	6.60%
Father	2	1.29%	9	6.08%	11	3.63%
Mother	7	4.52%	0	0.00%	7	2.31%
Grandfather	1	0.65%	6	4.05%	7	2.31%
Celebrity	2	1.29%	3	2.03%	5	1.65%
Family friend	2	1.29%	0	0.00%	2	0.66%
Sibling	0	0.00%	1	0.68%	1	0.33%
Family	1	0.65%	0	0.00%	1	0.33%
Not listed	111	71.61%	120	81.08%	231	76.24%
Total	155	100.00%	148	100.00%	303	100.00%

¹ Data indicates the total number of messages in which each individual was listed; some messages may list multiple individuals. Listed in order of total percentages.

). Of the 72 messages which specifically noted a namesake, the most frequently noted individuals were the child’s own older sister (27), followed by the child’s own older brother (20); unspecified older siblings were listed 1 time. As in Example 3, such messages did not necessarily state that they chose the same name, but rather that they sought names that were somehow connected to the child’s siblings. (Note that in Example 3, the number of strokes refers to the number of strokes needed to write the kanji; the selection of kanji for names is often influenced by fortune telling systems based on the number of strokes needed to write each kanji. See (Y. Kobayashi 2008) There were slightly more cases of children named for one of their grandparents (7) compared with the grandparents themselves being involved in the naming process (Example 4). In many cases naming children for others did not mean sharing actual names or parts of names, but being inspired by other people’s preferences, occupations or interests. One such case can be found in Example 5, whereby the connection was formed by using a kanji related to the father and grandfather’s work.

(3). 

Example 3.Futari no onēchan to no tsunagari no aru namae o kangae, jikaku o kangaete kazoku minna de kimeshita. [[We] thought of a name that was connected to [her] two older sisters, thought about the number of strokes in the name, and all the family together chose the name.] (Otobe General Coordination Department, Planning Section 2005b).

(4). 

Example 4.Meimei no hon o miteite, 薫乃 de ‘yukino’ to yomeru koto o shiri, daisuki na sofu no namae demo atta tame kimemashita. [[We] were looking at a naming book and learned that [you] could read ‘薫乃’ as Yukino, and it was also [my] beloved grandfather’s name so we chose it.] (Otobe General Coordination Department, Planning Section 2010d).

(5). 

Example 5.Otōsan ya ojīsan ga umi ni kakawaru shigoto o shiteiru no de, ‘umi’ to iu ji ga haitta namae o kangaete kimemashita. [[The child’s] father and grandfather, etc. work with the sea, so [we] thought and chose a name that included the kanji for ‘the sea’.] (Otobe General Coordination Department, Planning Section 2005a).

4.2. Commonalities Between Parents’ Names and Children’s Names

A little over half (167, 54.05%) of the children’s names generally used the types of non-standard readings that make them difficult to read (

Table 3. Orthographic characteristics of children’s names.

Reading Types	Child’s Gender
	Female		Male		Total
Non-standard readings or combinations 1	94	60.26%	73	47.71%	167	54.05%
Standard readings 2	62	39.74%	80	52.29%	142	45.95%
Total	156	100.00%	153	100.00%	309	100.00%

¹ Uses kanji readings that are not commonly recognized and commonly included in dictionaries, educational guidelines, etc., or combines readings in non-standard ways (e.g., mixes native Japanese and Sino-Japanese readings, etc.). ² Uses only kanji readings that would be commonly recognized and commonly included in dictionaries, educational guidelines, or uses the syllabaries.

), meaning the majority were of the new types of names said to be typical of contemporary naming practices. This also means that there were major differences between children’s and parent’s names, especially when matched in child-parent pairs—although this depends on what element is being examined. The number of children whose name used the same number of kanji as their father’s name was 215 (69.58%), compared with 165 (52.40%) who used the same number of kanji as their mother’s name (

Table 4. Number of paired names with commonalities between parents and children.

Parent	Kanji Location	Child’s Gender
		Female		Male		Total
Father’s name	Same length 1	102	65.38%	113	73.86%	215	69.58%
	Left 2	0	0.00%	0	0.00%	0	0.00%
	Middle 3	0	0.00%	0	0.00%	0	0.00%
	Right 4	0	0.00%	5	3.27%	5	1.62%
	Changed 5	0	0.00%	3	1.96%	3	0.97%
	Total	0	0.00%	8	5.23%	8	2.59%
Mother’s name	Same length 1	85	54.49%	80	52.29%	165	52.40%
	Left 2	1	0.64%	0	0.00%	1	0.32%
	Middle 3	0	0.00%	0	0.00%	0	0.00%
	Right 4	3	1.92%	0	0.00%	3	0.97%
	Changed 5	3	1.92%	0	0.00%	3	0.97%
	Total	7	4.49%	0	0.00%	7	2.27%
No common kanji		142	91.03%	137	89.54%	279	90.29%
Total		156	100.00%	153	100.00%	309	100.00%

¹ Both child and parent’s names use the same number of kanji. ² Both child and parent’s names use the same leftmost kanji. ³ Both child and parent’s names use the same middle kanji. Only relevant for names three or more kanji in length. ⁴ Both child and parent’s names use the same rightmost kanji. Only relevant for names two or more kanji in length. ⁵ Both child and parent’s names include a common kanji, but the location is different.

). This may be a case of conscious matching, but given that there are specific, strong patterns in the number of kanji used in names (Komori 2002; Unser-Schutz 2016b), with a tendency in particular for both men’s and women’s names to be between one and three kanji, this may not be intentional: 78.55% and 84.11% of girls’ and boys’ names, respectively, were written with two kanji, and none were written with more than three. That it was less common for girls to have the same number of kanji as their mother’s names (54.49%) than for boys to have the same number of kanji as their father’s names (73.86%) (Χ²(1) = 12.59, p = 0.000388) may also support this, as girls’ names were also more likely to use non-standard readings or combinations (Χ²(1) = 4.89, p = 0.026954), indicating that girls’ names are more frequently in the new style and consequently less like their mother’s names.

There were comparatively few commonalities in the orthographic characteristics of children’s and parents’ names. Only 15 children had names which used a kanji from their father’s or mother’s name. Of the eight children whose names had a common kanji with their father’s name, five used the kanji at the end of the name; three of the seven who had a common kanji with their mother’s names were also at the end. Being used at the end of the name can mean that it is one of the common name-exclusive suffixes (called tomeji); nonetheless, this makes the names look much more similar in form, and if the kanji are read in the same way, in rhythm as well. When the location of the kanji shifts in the parent’s name and the child’s name, it may be less clear that they are borrowing one of the kanji. Commonalities between parents’ and children’s names were only found between parents of the same sex, that is, all eight children who had commonalities with their father’s name were boys, and all seven children who had commonalities with their mother’s name were girls.

4.3. Commonalities Between Siblings’ Names

In total, there were 85 sibling groups. The largest sibling group consisted of four children, and a total of 193 of the children were found to have siblings in the data set (

Table 5. Number of paired names with commonalities between siblings (overview).

Number of Siblings in Set	All Female	All Male	Mixed 1	Total
2	23	15	26	64
3	4	4	11	19
4	0	1	1	2

¹Mixed indicates sibling set includes both girls and boys.

). It is possible that the 116 children not found to be in sibling sets also had siblings; since participation in the column is not obligatory, some children in the sample may have siblings who either did not have a message sent in, or whose message was sent in the period outside the current sample. Other sibling pairs may also be found in the remaining 116 children due to potential misspellings. For example, there were two children whose last name, area and father’s name were the same, but the mother’s name was listed as 由佳 and由香, two different ways of writing the name Yuka. In order to maintain an objective standard, such cases were not treated as siblings.

The majority of siblings were matched in terms of their orthographic length: Of the 85 sibling groups, 69 (81.18%) had the same number of kanji characters; an additional three sets of three or more siblings had some siblings with the same number of characters in their names (

Table 6. Number of paired names with commonalities between siblings (overview)

Shared Element	All 1		Partial 2		None 3		Total
Number of characters	69	81.18%	3	3.53%	13	15.29%	85	100.00%
Number of morae	50	58.82%	11	12.94%	24	28.24%	85	100.00%
Common phoneme	36	42.35%	16	18.82%	33	38.82%	85	100.00%
Types of readings used	21	24.71%	34	40.00%	30	35.29%	85	100.00%
Common kanji	15	17.65%	7	8.24%	63	74.12%	85	100.00%

¹ All of the siblings in the family shared the relevant element. ² Partial indicates some of the siblings in the family shared the relevant element, but not all. ³ None of the siblings in the family shared the relevant element.

). However, as with parents’ names, there are specific trends in the length of names observable over time (Komori 2002), making it problematic to interpret the meaning of the length of names. Similarly, 50 of the sibling sets (58.82%) had the same number of syllables, but the mora-length of names, too, is dependent on larger naming trends. Most importantly, 15 (17.65%) of the sets included a common kanji character; an additional seven sets of three or more siblings had some siblings with common kanji. Ten of the 15 sibling sets with common kanji matched on the final character tomeji suffix, making it the most frequent way to create commonality between names with kanji (

Table 7. Number of paired names with commonalities between siblings (detailed).

Element	Location	All 1		Partial 2		None 3		Total
Shared phoneme	Left 4	11	12.94%	7	8.24%	67	78.82%	85	100.00%
	Right 5	31	36.47%	13	15.29%	41	48.24%	85	100.00%
	Total	36	42.35%	16	18.82%	33	38.82%	85	100.00%
Shared kanji	Left 4	1	1.18%	1	1.18%	83	97.65%	85	100.00%
	Middle 6	1	1.18%	1	1.18%	83	97.65%	85	100.00%
	Right 5	10	11.76%	7	8.24%	68	80.00%	85	100.00%
	Changed 7	5	5.88%	1	1.18%	79	92.94%	85	100.00%
	Total	15	17.65%	7	8.24%	63	74.12%	85	100.00%

¹ All of the siblings in the family shared the relevant element. ² Partial indicates some of the siblings in the family shared the relevant element, but not all. ³ None of the siblings in the family shared the relevant element. ⁴ Sibling name used the same leftmost element. ⁵ Siblings name used the same leftmost element. ⁶ Sibling names used the same leftmost element. ⁷ Sibling names include a common kanji, but the location is different.

). An additional 36 sibling sets had phonetic commonalties (such has beginning with the same phoneme); 16 other sets of three or four siblings had some siblings with commonalities.

Unsurprisingly given that many siblings had matching final kanji, it was most common to have the same phoneme at the end of the name (31 sets). Many of these used the same characters, such as凌也 Ryōya and 祐也 Yūya; some, however, did not, such as 進之介 Shinnosuke and 健之助 Kennosuke. It should be noted that some of these similarities may be coincidental, given that Japanese prefers open syllables and only has five vowels, making for a limited number of name-final sounds. Although only two sibling sets included children who used the same kanji at the start of their name, 11 sibling sets started with the same phoneme, such as 怜子 Reiko and 竜輝 Ryūki and 果央 Kao and 季生 Kī. Given that many names are structured using a meaning-providing base kanji with a tomeji suffix—such as the typical girl names 花子 Hanako (flower+child) and 良子 Yoshiko (good+child)—using the same kanji to start a name may be too stark compared with a shared starting phoneme. Although the sharing of phonemes in different locations was not monitored due to the difficulty of ascertaining the likelihood of coincidence given Japanese’s limited set of sounds, an additional five sibling sets—and one partial set—had shared kanji in different locations. When the order changes, the commonality between the names becomes less obvious; but in some cases parents clearly sought to create connections between their children’s names while still maintain their uniqueness. For example, in one set of three sisters, the oldest daughter had the name 茉凛 Marin, and one part of her name was given to each of her younger sisters: 結愛 Yuma (commonly syllable: ma) and Rian 凛愛 (common kanji: 凛)—whose own names had matching kanji (愛).

5. Discussion

Given the role that names have in both creating ties within the family and raising consciousness about family relations, the above results show that by and large only the nuclear family—parents and their other children—are frequently involved in the naming process; it also suggests that family relationships outside of the nuclear family are not prioritized in the naming process. The fact that the children’s own older siblings are frequently stated as being involved in naming and that connections between the siblings are also created within the names indicates that an important role of contemporary naming practices is to strengthen intragenerational ties between siblings. Although naming practices in the pre-contemporary period played an important part in strengthening larger familial relationships through the creation of intergenerational ties and connections between parts of family clans, contemporary names appear to be playing an equally important role in the establishment of today’s smaller family unit.

Of course, it is impossible to say that because parents did not specify that grandparents or other family members outside of the nuclear family were involved in the naming process, they played no role therein. It is possible that grandparents and other members of their familial and local communities played more passive roles in the naming process, such as by expressing opinions about names. As Hendry (1989, p. 39) has observed, many nuclear families who express no interest in traditional rituals and activities such as shichiya—a celebratory feast on the 7th day after a child is born—or okuizome—a ritual held 100 days after birth to pray that the child never experiences food problems—still do so as they find it difficult to go against the larger community. Parents may feel similar social pressure with names, but do not feel it is significant enough—or a positive enough trait—to publicly note in such a short message. This is observable in the abundance of posts on the popular online forum Hatsugen Komachi ‘Small-talk Town’ asking for advice about how to negotiate family member’s (attempted) involvement in the naming process, such as Raito (2014) on how to deal with her father-in-law’s trying to choose her child’s name, or Sayuri (2018) on having matched names between cousins (for more information about Hatsugen Komachi see Unser-Schutz 2019).

Regional differences may also be at play. Although the types of names seen across Japan do not seem to differ by region (Unser-Schutz 2017a), the average household looks very different in different parts of Japan. Hokkaido also has one of the lowest rates of three-generation households across Japan (Raymo and Kaneda 2003), and as of 2005, only 18.85% of households in Otobe included three generations (Otobe General Coordination Department, Planning Section 2010a); in communities where three-generation households are more common, the results may differ somewhat. However, the tendencies observed in Otobe may be the future of Japanese naming practices: Even in communities where three-generation households are more common, they are still increasingly not the norm. In addition, given the brevity of these messages, parents must make decisions about what information to include. At the very least, it can be said that parents likely do not see the involvement of non-nuclear-family members as essential to the naming process.

Conversely, just because parents stated that a family member is involved does not mean that they played dramatic roles. Although several parents said that they involved their other children in the naming process, given that the older siblings were themselves young, it is likely in a relatively limited fashion. This is especially relevant for the selection of kanji: Instruction at school of the 1006 educational kanji is not finished until the 9th grade, and the full set of 2136 kanji for regular use is not completed until the 12th grade, meaning a very young child would not yet understand nuances of kanji. Although the older siblings may be involved in the selection process, it seems appropriate to assume that this is largely led by parents. This can be observed in the messages themselves, as in Example 6 and Example 7, which both point to a two-part process, whereby the older siblings select a phonetic form and their parents select an orthographic form. When older siblings are involved in selecting the orthographic form, they may also be selecting kanji from a limited set of kanji with personal significance, such as in Example 8, where the older sister gave the baby sister a kanji from her own name.

(6). 

Example 6.Onēchan ga namae o tsuke, ato wa kakusū o mite kimemashita. [[Her] older sister chose the name, and then [we] looked at the number of strokes and decided.] (Otobe General Coordination Department, Planning Section 2010c).

(7). 

Example 7.Yobikata wa onīchan ga kangaete, otōsan to okāsan ga jikaku o kangaete kimemashita. [[Her] older brother chose the phonetic form, and mother and father thought about the number of strokes and decided.] (Otobe General Coordination Department, Planning Section 2010b).

(8). 

Example 8.Onēchan ga, jibun no namae kara 1ji o toritsukemashita. [[Her] older sister gave her one kanji from her own name.] (Otobe General Coordination Department, Planning Section 2008).

Nonetheless, the fact that parents deem it noteworthy in these short messages again suggests that this is something that they want to emphasize. There may be a practical or strategic side to this: With the birth of a younger sibling, a child’s relationship with their parents and position within the family changes and can leave older children feeling left out. In the Japanese context, it is common for parents to give older children special roles by emphasizing that they are more experienced and understanding (Hendry 1989, p. 56). One can imagine that involving older children in the naming process similarly helps give them an opportunity to feel a special connection with their younger siblings; additionally, by describing the selection of names in this way, parents are able to create a narrative that strengthens intragenerational ties between siblings.

Given how dramatically different contemporary names are from those of previous generations, it is also important to consider how these changes in familial relationships have worked in tandem with social changes to allow the development of new types of names. Changes in familial relationships may license new types of names by reducing social pressure to select orthodox names, but people’s preferences for new types of names are likely developed by other factors. As noted in the overview, recent names seem to have been affected by changes in values, and in particular, a desire for uniqueness that seems to have fueled on creative choices in names, which have often led to their being difficult to read. Y. Kobayashi (2009) has argued that these changes can be attributed in part to the loss of a sense of the public; that is, individuals who are generally less conscious of the public sphere may be less likely to consider the burden that difficult names present to others who are a part of the greater public.

Relating this to familial relationships, the fact that only members of the nuclear family are involved in the naming process today may contribute to feeling that it is a largely private activity. In so far as a smaller number of actors with immediate relationships—both between themselves and the child being named—are involved, the act of naming may seem to be more intimate than it did for previous generations. How parents seek to create relationships through names may support this argument. As noted in the analysis of Example 5, even when connections are made by naming children after other family members, it is often in a way that is less obvious and more private in nature. In comparison with naming a child for someone through the use of shared phonological or orthographic characteristics, in cases like Example 5, the fact that the child has been named after a family member would not be obvious to people who lack personal information about the family beyond their names, making this a highly intimate—and private—way to create bonds.

It is interesting to note that this growing intimacy of naming practices comes at the same time that people are becoming sensitive to how personal information and names are regulated and shared. Through the increasing regulation of individuals through their names in the vast and interrelated governmental and pseudo-governmental network of personal data (birth registration, family registers, My Number personal identification numbers (the Japanese equivalent of the US’s social security number, first implemented in 2015), school records, credit cards, bank mortgages), names are used and recorded in a variety of spaces outside of an individual’s own control. With the rise of the internet, the sharing of personal data is a daily occurrence for many individuals. Although proving this is beyond this article’s scope, increasing the intimacy of the act of naming by limiting those involved in the process may be potentially read as an act of resistance towards the lack of control people experience over their own identifying data.

6. Conclusions

Names not only identify individuals, but as Bodenhorn and vom Bruck (2009) have noted, they also embed individuals into a matrix of relationships with others. As this article has shown, changes in naming practices consequently often reflect changes in familial relationships; in the case of contemporary Japan, the emergence of the nuclear family as the core familial structure seems to coincide with a tendency for parents to use names as a way to strengthen relationships within the nuclear family. As the data used in this study were comparatively brief messages, it is possible that other information affecting the selection of names was limited due to space concerns; as such, it should not be read as fully reflecting all aspects of contemporary Japanese naming practices. Given that the process of choosing a name that takes place over several months—for many parents, at the start of pregnancy if not before, to 14 days after birth when the birth certificate finalizing the child’s name must be submitted—it is something that can be both a worrisome (e.g., thinking ‘Are we choosing a good name?’) and pleasurable experience that is hard to describe in two lines.

Brief messages do, however, have the benefit of being highly focused: This is the information that parents chose to present as their priorities, and in that respect, it is clear that only the most immediate of family members—that is, the newborn’s parents and siblings—are seen as having noteworthy enough roles in selecting names. This data also has the advantage of being fully public, and the fact that it includes the naming information of both children and their parents—which made the current study possible—is both valuable and difficult to achieve using traditional surveys or data sets. Confirming the patterns observed here is a pressing task and increasing the data set—both through the potential collection of further back issues and continuing to add data as new issues are published—is one way to do so. In addition, future research would be well-served by increasing the data used, particularly to include areas outside of Otobe in order to confirm the generality of the data reported here. Although the current space limitations do not allow for further analysis, I have conducted a large scale survey concerning parents’ reasons for selecting names, which will be reported on separately.

Another potentially fruitful avenue of research may be to examine how parents negotiate dealing with family members who seek to influence the naming or provide advice. Some anecdotal evidence suggests that the naming process can create conflicts within families, in large part due to differences in opinion about who should be involved in naming and what is an appropriate name. Look, for example, at discussions online such as Mana (2014), who sought advice on Hatsugen Komachi after her own mother pressured her to select a different name from the one she had chosen for her child, or the examples of Raito (2014) and Sayuri (2018) above. In addition to the quantitative analyses already in motion, qualitatively analyzing how families deal with conflicting cases such as Mana’s, Raito’s and Sayuri’s may offer insight into both how familial relationships affect naming practices and, conversely, how the experience of choosing names affects familial relationships themselves.