For all the passion their creators poured into them, pamphlets and paintings were relatively cheap. Th
e perceived urgency of fi ghting infanticide is more impressively evinced by the prodigious resources that governments and individuals poured into far costlier countermeasures. If moral suasion redrew the boundaries of humanity and argued that infanticide was not compatible with a pleasant aft erlife, childrearing subsidies and pregnancy surveillance focused on the material conditions that their designers saw as motivating and permitting infanticide: poverty and impunity. Such policies reached into remote villages, consumed great quantities of paper, and tied up considerable administrative and fi nancial resources. Depending on the locality, a sizable fraction of parents received subsidies for their children.
Pregnancies were registered in the hundreds of thousands. Offi cials and doctors
investigated stillbirths and infant deaths, and on occasion jailed men who had not bothered to report their wife’s pregnancy.
In spite of their scale, the evidence for the ultimate success of subsidies and surveillance is surprisingly equivocal. Infanticide declined where child bounties were paid and pregnancies monitored, but this decline was neither complete nor did it always outpace the same trend in areas without such policies. Th is may suggest that the true power of subsidies and surveillance lay not in their eff ect on the relative costs of childrearing and infanticide but in their symbolism. When lords rewarded parents with coin and rice from their privy purse, or when local notables made large donations to subsidize the childrearing of their less pecunious neighbors, they elevated the social meaning of childbirth. Placing pregnancies under surveillance stated plainly that they were not a private matter but the concern of the state. Bearing and rearing children, therefore, was no longer a matter for 158
17/04/13 3:54 PM
Subsidies and Surveillance 159
individual families or even individual villages. Every child was a contribution to society—to a village forlorn amid its fallow fi elds, to a lord over disappearing taxpayers, perhaps even to the whole of Japan. Th
e expensive policies powerfully
embodied this notion; once it took root in the minds of villagers, their conversations, friendships, and marriages bore it beyond the borders of their domain.
H O W SU B S I D I E S A N D SU RV E I L L A N C E C A M E T O B E
E X P E C T E D F E AT U R E S O F G O O D G OV E R NA N C E
It was at the very edge of the archipelago, in the island domain of Tsushima mid-way between Kyushu and Korea, that Japan’s fi rst childrearing subsidies were disbursed in 1667.1 Since the domain received a large grain allowance from Korea, this may have been a way to share some of that bounty with peasants struggling to coax barley from their windswept rocks.2 Every peasant child, the lord of Tsushima announced, would receive one bale of grain for each of the fi rst three years of its life. While subsequent reforms made fi rstborn children ineligible and changed the amount and timing of the payout, Japan’s earliest childrearing subsidy regime was also one of its most inclusive. Between 1683 and 1701, 4,417 children, or about every third child born during those years, received a total of 8,834 bales of umaregomugi (“barley for babies”) *. * 3
Th
e next time we hear of policies to encourage childrearing, they issue from the center of the Tokugawa polity. In 1690, Tokugawa Tsunayoshi, who over the previous years had banned the maltreatment of dogs, horses, and nesting birds, ordered that people too poor to raise their children should appeal to their employers, headmen, or magistrates for help. Pregnant women and children under seven would henceforth require registration.4 Another law later ordered tenants in Edo to report pregnancies through their landlords, and each ward to forward a list of infants to the town magistrate. It is not clear how long this system operated in Edo, however, and it is unlikely that anything like it was ever implemented in the shogun’s villages.5
In emulation of Tsunayoshi’s laws, the domains of Kaga and Shinjō ordered the registration of pregnancies in the mid-1690s, and Owari promised economic support for abandoned children.6 Although the surviving documentation is modest, Kaga seems to have continued its policies against child abandonment for generations. When in 1767 the shogunate asked all domains to forward a ban on infanticide, Kaga responded that newborns were already safe within its borders.7 Shinjō
repeated and elaborated its laws in 1718 and 1733, but if pregnancy surveillance records were actually compiled at this time, they have not come down to us.8
Aft er his death, Tsunayoshi was widely remembered as an eccentric. For the next two or three generations, few people worried about protecting newborns from their parents.
17/04/13 3:54 PM
160 Redefining Reproduction
Aft er half a century of low fertility and incipient population decline, infanticide fi nally emerged as a major policy issue in the 1740s, when the fi rst domains of Eastern Japan introduced childrearing subsidies. In Shōnai, a domain elder is credited with dispensing “childrearing rice” around 1740.9 In Nihonmatsu, a senior councillor convinced the young lord to launch a fi nely graduated system of childrearing grants in 1745, and village records from that decade show that pregnancies were monitored.10 In 1751, the lord of Shinjō began to pay bounties as well.
His generosity did not produce large families overnight: When in 1765 he rewarded parents of at least fi ve children with one bale of rice, only 123 couples qualifi ed in a population of nearly 50,000.11
Such policies were still too extravagant for many tastes. In 1745, the castletown magistrate of Aizu, Kamio Taizō, petitioned his superiors to disburse rice to all children with older siblings if their parents were too poor to employ servants.
Kamio planned to raise a part of the necessary funds by fi ning anyone who killed their fi rst child. Th
e main source of fi nancing, however, points to the institutional inspiration for childrearing grants and endowments. Kamio wanted to partially convert a long-standing relief measure—rice grants to the poor—and to rededicate the purpose of a more sophisticated piece of fi nancial infrastructure, the castletown’s fi re-and-fl ood endowment, whose grants were fi nanced from the 100 ryō it generated in interest each year. Kamio’s scheme was taken seriously enough to be discussed in a conference of the lord and his councillors. However, they turned it down for lack of precedent.12 Five years later, in 1750, Suzuki Heijiemon, assistant to a district magistrate, was no more successful with a similar petition in Mito.13 In Sendai, a retainer of rank proposed childrearing subsidies in the 1760s; while a number of village granaries subsequently made loans to indigent parents, this does not seem to have been a domain-wide system.14 One by one, other governments announced childrearing subsidies and pregnancy surveillance.15 In 1776 and 1778, Aizu and Mito, both ruled by new lords, introduced tentative childrearing grants.16 Nonetheless, even in 1780 only a small minority of Eastern Japan’s impoverished parents lived under regimes that gave them offi cial
assistance.
Th
en the famine of 1783 and 1784 created an urgent sense of demographic crisis.
By the end of the 1780s, the population of most major domains in both the Northeast and the North Kantō had fallen at least a quarter below its peak value. To many observers, the unmistakable signs of commercial growth amid demographic decline only underscored the moral crisis.17 Within years, the number of surveillance and subsidy regimes doubled. Th
ree of the largest domains of Eastern Japan—
Mito, Aizu, and, by 1807, Sendai—replaced piecemeal policies with comprehensive systems. Some domains that had granted subsidies for decades made them more generous in the 1780s and 1790s. Nihonmatsu, for example, made second children eligible for grants in 1786, doubled the allowance for third and fourth children, and
17/04/13 3:54 PM
Subsidies and Surveillance 161
tripled it for fi ft h children. In 1797, it raised these amounts by another third.18 In the same period, a new cohort of intendants ( daikan) announced infanticide countermeasures in the shogunal territories they administered. Between 1789 and 1797, four intendancies launched such policies, followed by another four by 1814.19 Aft er years of eff orts to revitalize their lands, some of these intendants became local gods, with shrines dedicated to their veneration, and others had tearful commoners beg them to stay when their appointments ended.20
Hirose Ihachirō’s tenure as intendant at Kashiwagura in Dewa was too short to leave such an impression on the locals. He did, however, leave a trail of pleased local offi
cials on his progress from Edo to his new appointment. It was the winter of 1789.
Before his departure, he talked to colleagues in Edo and inquired about the child welfare policies of various domains. Straddling Hirose’s road to the north lay Kurobane, where childrearing subsidies had been disbursed for twenty-seven years, and where Suzuki Busuke had distributed his poster with the cat-faced woman a few months earlier. Two letters from Hirose survive in Suzuki’s papers, but they did not meet on this occasion, as Suzuki was in Edo on offi
cial business.21 At the Kurobane
inn where he spent the night, Hirose instead questioned the offi cial in charge of his
lodgings about the governance of the domain lord. As local headmen later recalled with pleasure, Hirose requested a copy of Kurobane’s decrees against infanticide.
When he later returned to Edo, Hirose apparently read them to his colleagues and then formally returned them to Kurobane’s Edo mansion.22
Aft er the stop in Kurobane, Hirose continued his northward journey. Aft er several days, he reached Nihonmatsu, where he interviewed a local administrator, Shidara Sukezaemon, about Nihonmatsu’s child welfare policies. Shidara made a point of impressing Nihonmatsu’s pioneering role on his visitor, then sent Hirose on his way with a copy of the original laws of 1745—even though they had been revised several times and just recently been made much more generous.
Infanticide countermeasures had evidently become a source of local pride.
Before long, they became an expected feature of good government. In this spirit, one of Hirose’s fellow intendants, Teranishi Jūjirō, invited ten Eastern domains to two conferences on the eradication of infanticide, in 1811 and 1817
respectively.23
Where governments did not take the initiative, villagers frequently petitioned for infanticide countermeasures or even spontaneously off ered money.24 Some succeeded in enlisting the institutional support of their governments. For example, Zenbei, the headman of Shimotaki in Kōzuke, fi rst petitioned his intendant in 1785 to accept 100 ryō and in return give Zenbei 10 ryō per year to distribute to poor parents in his village for childrearing purposes. Th e intendant evidently
accepted the off er, and the system remained operational under his successors until at least 1820. Other headmen followed Zenbei’s example, such as Shinbei of neighboring Narahara, who off ered 50 ryō in 1791.25
17/04/13 3:54 PM
162 Redefining Reproduction
Elsewhere, however, especially on the Bōsō peninsula and in northern Musashi, concerned individuals failed to convince their governments of the necessity of population policies.26 For example, Hirayama Chūbei, an important landowner and village headman in eastern Shimōsa, wrote a petition to his lord, a minor bannerman. In the provinces of Kazusa and Shimōsa, Hirayama lamented, the population was dwindling, fi elds were reverting to wasteland, and households were going extinct because people would raise no more than three children. “If we can stop this one thing, the population will increase year by year . . . and abandoned lands will naturally be opened to cultivation and must ultimately redound to a vast profi t.” In an attachment, he explained how infanticide could be stopped: As much as we would like to help the poor in perpetuity, the likes of us rise and fall quickly and it is unlikely that we will be able to help them forever. Aft er considering many ideas, we thought of the following solution. We would devote ourselves to our household business, economize on expenses, and save what was left . Once our savings reached an appropriate magnitude, we would humbly give them to His Lordship, and beg him to take them as a credit [in addition to the money he already owes us]. From the interest we would receive every year, we would like to help the poor in perpetuity.27
Unfortunately, Hirayama explained, his father had died before they could implement the plan; however, he had left behind the stately sum of a thousand ryō, 500
of which he now off ered as a loan to the bannerman. He did not mention the interest rate, but judging from his plan to pay 5 ryō to every poor child in his village, he was expecting a good return. Th
e bannerman, however, was unswayed
by Hirayama’s protestations of how grateful his late father would be as he lay “in the shadow of grass blades.” Perhaps saving infants and stopping the depopulation of his villages was a smaller concern to the bannerman than that of taking on a large debt. In reply, he sent Hirayama an Edo-period standard rejection letter: there was no precedent for this commendable scheme.28
Where domains were not ready to make childrearing subsidies an offi cial policy, wealthy commoners frequently stepped in with their own funds. One of the more extraordinary such eff orts unfolded in western Honshu, where a father-andson pair of wealthy commoners supported the parents of no fewer than 110 children in eight districts of several domains.29
Nevertheless, domains remained the most important backers of such policies, and more and more of them introduced subsidies and surveillance in the nineteenth century (Figure 17). Many ruled lands in the North Kantō and the Northeast (Nanokaichi, Numata, Maebashi, Moriyama, Sakura, Oshi, and Ushiku), but subsidies and surveillance also spread into Obi in southern Kyushu and a small part of Shinano (Komoro, Iwamurata, and some shogunal lands).
Almost all established programs remained in place. In some cases, this required an
17/04/13 3:54 PM
60
50
40
30
20
10
0
Subsidies and Surveillance
Surveillance only
Subsidies only
figure 17. Number of domains and intendancies that monitored pregnancies and granted childrearing subsidies, all of Japan, 1650–1870. (Th
e chart subtracts offi
cially announced
discontinuations, but not systems whose enforcement grew lax. It excludes the island kingdom of Ryūkyū, which was not formally part of Japan when it announced subsidies and surveillance in 1750. sources: See Appendix 7.)
17/04/13 3:54 PM
164 Redefining Reproduction
extension beyond their original expiration dates.30 Some systems became less generous and probably also less rigorous, and at least one was dismantled altogether.31
Most, however, continued to consume great administrative and fi nancial resources until the abolition of domains in the late 1860s and early 1870s.
T H E F I NA N C E S O F B E N E VO L E N C E
Subsidies were expensive. Since it was a rare domain that was not mired in debt by the late eighteenth century, subsidies were not usually fi nanced from the regular budget. In some domains, members of the ruling family opened their privy purses.
For example, a lady of Sendai donated her dowry, a lord of Fukuoka curtailed his personal expenses, and an Aizu lord bequeathed his lifetime savings of 1,500 gold ryō to a childrearing fund.32 Th
e shogunal intendant Teranishi Jūjirō was able to
borrow 5,000 ryō from the shogunate, loaned the money in turn, and fi nanced his childrearing scheme from the interest.33 His colleague Kishimoto Takedayū
obtained a childrearing fund of 10,000 ryō from the shogunate but grumbled that this was less than a third of what the neighboring intendant, Takegaki Naohiro, had received.34 Far more frequently, however, it was rich commoners who supplied the funds for the childrearing subsidies.
For particularly wealthy commoners, paying the poor not to kill their children was part of a larger system of charity that kept them on good terms with their neighbors and their lords.35 One case in point is Sekizawa Masahide, a rural merchant of Mito who responded with a gift of 100 ryō when his lord announced childrearing subsidies in 1791. Masahide had inherited a fortune of 5,000 ryō in 1776, built primarily on exporting paper and brewing soy sauce and saké. When he retired forty years later, it had grown to a princely 20,000 ryō and his business network extended to fourteen provinces. When a fi re caused damage in 1812, fourteen condolences arrived from Edo alone, and another four from Osaka and Kyoto. In this sense, Masahide exemplifi ed the ways in which commoner elites moved in circles far beyond the world of their village and even their domain.
Maintaining good relations with both was nonetheless important. From Masahide’s papers, which include an elaborate policy proposal, it is clear that he was concerned with rural revitalization. He also believed in charity. In instructions he wrote late in life for his daughter and heir, Sei, he urged her to emulate his devotion to profi ts because it would enable her and his son-in-law to provide for their descendants, serve their lord, and help the poor. To this day, the Sekizawa own a tripartite chest that has one coin slot for “contributions to the fi sc,” one for “provision for descendants,” and one for “friendly succor to the little people.”36
In Masahide’s day, such succor included interest-free loans of 1,092 ryō to people in twenty-nine villages between 1783 and 1803 and, in 1808, the construction
17/04/13 3:54 PM
Subsidies and Surveillance 165
of a brand-new storehouse stocked with 30 ryō and 300 bales of millet for his home village. Such conspicuous philanthropy evidently smoothed the family’s relations with their less fortunate neighbors, and for decades the Sekizawa never suff ered one of the “smashings” that the unpopular rich of the late Edo period had to fear.37
Sekizawa Masahide also parlayed his generosity into a privileged relationship with Mito’s ruling family. When Lord Harumori entertained four hundred of his commoner subjects to recognize their contributions, Masahide sat in a place of honor. When Masahide later paid a visit to the lord’s wife, she bounced his infant grandson on her lap. Many years later—long aft er Masahide’s death—the lady dropped in on the Sekizawa during a mushroom hunting expedition. Th e Mito
lord himself sent a poem to Sekizawa’s grandmother to celebrate her longevity. In an age when fortunes could be lost as quickly as they were made, such tokens of lordly aff ection were priceless.38
Over the years, the Sekizawa were repeatedly off ered warrior status in recognition of their frequent contributions to Mito’s fi sc. While they declined the honor, other philanthropists eagerly accepted this reward. In Nihonmatsu, the expansion of the subsidies in 1786 was only possible because private benefactors volunteered their donations. One of these was Urai Heibei, a brewer and pawnbroker of Motomiya. He donated 100 ryō that year, and another 1,000 ryō in 1802.39 In 1803, he received warrior status along with a fi ef of 100 koku. Urai’s descendants continued to benefi t from the special status his philanthropy bought for the family. In 1815, Heibei’s son Heizaemon was able to borrow 200 ryō from the funds of the childrearing system at 6.125 percent per annum, half the usual rate of interest.40
Others gave more modest sums, but from fortunes that were themselves unspectacular. Oft en, they participated in subscriptions. Some such funds were raised in individual villages; others were collected for an entire domain, as in Numata, where three energetic priests mounted a formidable fundraising eff ort in 1818.41 Donors were oft en carried along by their peers. For example, the Akita townsman Nanami Saburōemon and his friend Shioya Zenbei rallied a group of 191 donors whose contributions ranged from 5 to 30 ryō and added up to more than 2,000 ryō.42
Domains frequently prodded the charitable impulses of their subjects. Nanami’s eff ort began in a conversation with his magistrate, who mentioned that their lord was hoping to expand subsidies from parts of the countryside to the castletown. A bannerman appealed to his two villages in Kōzuke with the following words: “Since our residence is reduced to strict austerity and we nevertheless grant this special munifi cence, let men of compassionate hearts help shoulder the burden of the childrearing subsidies. Th
eir names will be recorded so that their
glory will be known to posterity.”43
17/04/13 3:54 PM
166 Redefining Reproduction
In the Minamiyama Okurairiryō, a shogunal territory administered by Aizu, the domain did not rely on the generosity of its more fortunate subjects alone.
Instead, it divided the population into ten wealth strata. Beginning with the
“upper lower class” ( ge no jō), people were assessed half a ryō per household. Four strata higher, the “lower upper class” ( jō no ge) had to shell out 7 ryō; and households designated “particularly upper class” ( betsujō) had to pay a heft y 50 ryō. As a result, individual groups of villages with about two thousand inhabitants ultimately managed childrearing endowments of 250 or 600 ryō.44
Th
e fi nancial sophistication of the childrearing funds is notable. Once the funds were collected, they typically functioned as an endowment that preserved the principal and covered running expenses from interest alone. By the mid–
nineteenth century, there must have been several hundred such endowments.
Some were large funds administered by a domain or intendancy, while others were more modest aff airs run at the village level. Th
e endowment would be invested in
loans—oft en to the domain itself, but also to rural entrepreneurs—or mutual savings groups, typically with an interest rate of 10 to 15 percent.45
An ingenious variant of the childrearing endowment was the Silkworm Deposit Fund ( kaiko tsumikin) of Maebashi domain. From 1848, Maebashi collected a percentage of the revenue from cocoon sales in this fl ourishing sericulture region. ** **It retained these funds for fi ve years, then returned the principal to the sericulturalists while disbursing the interest as childrearing subsidies.46
Th
e system was apparently considered a success; in 1859, neighboring Numata copied it.47
T H E S C A L E O F T H E SU B S I D I E S
With the interest from their endowments, some regimes were able to pay grants to a majority of local children (Table 6). In one ward of Kōriyama in Nihonmatsu, twelve children were born between the fi ft h and the ninth months of 1815. One of these children had no older siblings and therefore did not qualify for a subsidy, and one died within its fi rst month. Of the remaining ten, only one child, born to a wealthy family that was promoted to rural warrior ( gōshi) status that year, did not receive a subsidy. Th
e remaining nine children were all paid between half a ryō
and 3 ryō, including the fourth child of a couple that owned more than 20 koku of land and was suff ering no want.48 Th
e village of Onna in Hitachi has records of
subsidies for over forty years.49 In an average year, 40 of its ca. 115 households received subsidies. Around 1830, that number reached 75, suggesting that every new child in the village received a grant. More typically, subsidies were limited to the poor, the prolifi c, or the prolifi c poor. Th
e scale of the payouts in such regimes
varied widely, but even at the low end of the range, about 4 percent of children benefi ted from a grant.
17/04/13 3:54 PM
Subsidies and Surveillance 167
table 6 Percentage of newborns who received childrearing subsidies in various jurisdictions Subsidized
births
Location, in descending
As % of
order of generosity
Years
No.
all births
Kotagawa in Shirakawa domain
1793
12
Close to 100
Onna in shogunal territory
1791–1832
1,633
Close to 100
Nakaishii in the Hanawa Intendancy
1802–1811
ca. 300
Close to 100
Kōriyama Kamimachi in Nihonmatsu domain
1815
9
ca. 70
Tsuda and Kusuriya in Mito domain
1808, 1810, 1812,
45
63
1834–1837,
1841
Tsushima domain
1683–1701
4,417
36
Shogunal subjects in nine villages in Kōzuke
1839
29
30
Minamihase in Sendai domain; loans only
1796–1797
12
29
Twelve villages in Minamiyama Okurairiryō,
Eleven years between
135
27
shogunal lands under Aizu administration
1817 and 1836
Oguki in Ushiku domain
1851–1870
25
ca. 22
Ihara and neighboring villages in
1807–1808, 1820,
Fukuoka domain
1831–1832
48
17
Namashina in Numata
1844
3
15
Ogachi and Hiraka districts in Akita domain
1825–1830
837
9
Shinjō domain
1751
120
9
44 villages in Miharu domain
1788–1818
439
4.5
sources: For all locations, the total number of births was either reported in the same document or estimated based on births in other years or on population size and an assumed birthrate of 25 per thousand. Kotagawa: Shirakawa *shishi 7, * 387–388; rate based on 1825 population of 490 and 14 births in 1830. Onna: Yamakawa, 42-ga-nenbun shōni onteate watarikata torishirabechō (1833); Kōriyama Kamimachi: Takahashi, *Zaigōmachi, * 134–135. In the sample at hand, 81 percent of children received bounties, but if the same rules applied in an average year, only about 70 percent of newborns would have had older living siblings, a prerequisite for receiving the subsidies. Nakaishii: Based on the expenditure on grants; in Fukushima-ken Kyōiku Iinkai, eds., *Teranishi daikan-ryō, * 197–198. Tsuda: *Katsuda shishi Chūsei-hen Kinsei-hen, * 852. Kusuriya: *Kanasagō sonshi, * 555. Oguki: *Ushiku shishi Kinsei, * 384–385. Tsushima: Hironaga, “Tsushima-han,” 34. Shogunal villages in Kōzuke: *Iwashima sonshi, * 760–761. Minamihase: Ohara, “Sendai-han,” 98–99. Minamiyama Okurairiryō: Matsueda, *Aizu-han, * 102–103, 123; denominator estimated based on 46 births in a population of 2,594 in 1827. Ihara district: Kalland and Pedersen, “Famine and Population in Fukuoka Domain,”
- Namashina: *Gunma kenshi shiryōhen 12, * 431. Ogachi and Hiraka: Takahashi, *Nihon jinkōshi 2, * 160. Shinjō domain: *Shinjō shishi 3, * 42. Miharu villages: documents in *Miharu chōshi 9, * 566–573. Th e estimate for total births is based
the 1719 headcount of the 44 villages, reduced by the one-third population decline for all of Miharu between 1719
and 1786.
Relying on such cases alone, picked simply for their surviving documentation, of course overstates the extent of childrearing subsidies. In villages and years in which no subsidies were paid, no such documents would have been generated. Th e
largest domain of Eastern Japan, Sendai, may have been among its least generous.50
Still, it is not fanciful to think that at least 10 percent of Eastern Japan’s children received childrearing subsidies. Given that most fi rst and many second children did not qualify for subsidies, the proportion of third or fourth children who enjoyed offi
cial support would have been much higher still.
17/04/13 3:54 PM
168 Redefining Reproduction
No two domains were alike in the precise rules of eligibility and the amount and denomination of the subsidy. Only poor parents qualifi ed for aid in some systems, while elsewhere, the rules did not discriminate based on need. Some domains disbursed the same bounty for every child aft er the fi rst, but some paid more for higher-parity children. Nihonmatsu’s grants for a fi ft h child were more than fi ve times as generous as those for a second child. Miharu briefl y paid four times the usual subsidy to children born within a year of a sibling, and Yonezawa reserved its greatest largesse for the rare couples that had fi ve children under the age of sixteen. Th
e timing of the payout also varied. Fukuoka granted a bale of rice already before the end of the pregnancy, but Hanawa and Takada only awarded subsidies to children who had survived their seventh night. Possibly this was a recognition of the concentration of infant mortality in the fi rst few days aft er birth, but the rule may also have refl ected popular notions of important milestones on the road to full human status. Th
e generosity of payments also
varied. At the low end, parents received a one-time grant of a bale of rice or half a gold ryō. More typically, however, domains disbursed grants of a similar magnitude annually for the fi rst three years of the child’s life. At the high end, the benefi ts could be very substantial. Th
e Mōka intendancy paid monthly install-
ments totaling 7 ryō over fi ve years, and in Sōma Nakamura domain, villagers with three sons or three daughters received nine bales of rice over seven years.51
How far would individual subsidy payments go? As we saw in Chapter 5, raising a child to the age of three without extravagance cost a little less than 5 ryō in food, clothes, and birth-related expenses. If we assume that childrearing diminished a woman’s ability to work by about a third—an arbitrary but plausible number—and that a woman’s labor was worth about 4 ryō per year, the outlays and lost income of raising a child to the age of three would have amounted to about 9 ryō.52 Only the most generous systems came close to off setting this cost completely, but most paid some considerable fraction of this expense.
Where subsidies were meager, they were not above criticism. In Shimōsa, a retired merchant petitioned his intendant to increase subsidies from a quarter ryō, which he thought quite inadequate. Pointing to the generosity of the neighboring intendacy of Mōka, he suggested 4.5 ryō over the fi rst three years of a child’s life as a more appropriate sum.53 In one Shimotsuke village, local philanthropists took matters into their own hands; in the rules of their childrearing foundation, they promised the poor 500 copper mon per month during pregnancy and a total of 9
ryō aft er the child’s birth.54
In all systems, new parents had to fi le applications to receive a grant. Fukuoka made this easy, as every pregnancy report included a line on whether the couple was requesting a subsidy.55 More typically, the application involved considerable paperwork. Frequently, applicants had to demonstrate that they could not raise a child unaided, as in the following example from Sendai:
17/04/13 3:54 PM
Subsidies and Surveillance 169
A humble request for a subsidy for raising an infant
For Kyūemon of [the village of] Fujisawa Hongō in Higashiyama Landholdings of 31 mon [roughly 0.3 koku] and 35 mon in newly reclaimed land Kyūemon, [resident at the compound of] Shinchi Yashiki Ondōbori Kyūemon
42 years of age
Wife, Mase
32 years of age
Daughter, Fuyo
12 years of age
Second son, Seiemon
8 years of age
Th
ird daughter, Iyo
5 years of age
Fourth daughter, Chō
born this year
6 people in total, 2 males and 4 females
With landholdings and family size as noted above, Kyūemon has for a long time truly been a man of the righteous road, and has applied himself vigorously to agriculture. However, since what little land he owns is of poor soil, he is destitute and cannot support his family. We therefore humbly request that you investigate this case and, in recognition of his situation, grant a subsidy. We make this request with our joint signatures.
Head of the fi ve-man group
Seinosuke
10th year of Tenpō [1839], 9th month
Childrearing Commissioner for this village
Kishirō
and head of a fi ve-man group
Childrearing Commissioner
Kiheiji
Acting headman of this village
Oikawa Hōichirō
To the Supra-Headman ( Ōkimoiri) Iwabuchi Tōshichirō56
Accepting a subsidy entailed obligations. On a summer’s day in 1830, the 837
children that had been “saved” by a particularly energetic magistrate, Kuniyasu Matazaemon, gathered with their parents on the grounds of a temple in rural Akita. Th
ere they received a breakfast and lunch. When the male participants in this picnic turned twelve or thirteen, Kuniyasu invited them to come to two special villages and gave them land to till. So far, so good, but such children were known as *onbuikunin, * or “people nurtured by His Lordship’s charity,” even aft er they grew up. When the Akita lord held his processions to and from Edo, the onbuikunin had to line the roads in a display of gratitude. By the late 1840s, some onbuikunin had achieved personal wealth or married into well-established families; many found it embarrassing to have to present themselves as former welfare cases. Kuniyasu, now an old man, therefore caused these ceremonies to be ended.57 A distaste for the humiliation of collecting welfare may also explain why aft er Annaka domain introduced subsidies in 1808, it found that not one person had fi led an application.58
P R E G NA N C Y SU RV E I L L A N C E
While in most domains subsidies aff ected only a minority of parents directly, no commoner household was exempted from pregnancy surveillance. Th e paperwork
17/04/13 3:54 PM
170 Redefining Reproduction
was extensive, requiring a minimum of a pregnancy report and birth report. Ordinary villagers rarely had the literacy required for writing the sure hand and formal language of offi
cial documents. Th
e introduction of pregnancy surveillance there-
fore further increased the considerable administrative burden of village elites. Th e
form of the initial pregnancy report varied from domain to domain, but the following document from Numata is typical:
I respectfully report in writing:
My wife, Waka, aged 26 in this year of the dragon, is pregnant, and we expect her to deliver around the 8th month. Once delivery is imminent, the neighbors and the fi ve-man group will report this to the village offi
cials. [My wife] will give birth once
they have assembled. In the unlikely case that the birth will be diffi cult, the village
offi
cials will make a report to the Childrearing Commissioner. If the pregnancy ends in a miscarriage ( ryūzan) or something of that sort and we fail to report this, we will not object to any punishment that His Lordship sees fi t to impose on us.
Namashina village
Report fi
led
by Den’uemon
15th year of Tenpō [1844], 3rd month
To the Offi
ce of the Childrearing Commissioner
Confi rmed by
Th
e Headman
Th
e Head of the fi ve-man group59
Just as he had promised, Den’uemon fi led a birth report, but sooner than expected.
I respectfully report in writing:
Den’uemon [seal]
His wife, Waka [seal]
aged 26 in this year of the dragon
When [Waka] hoed the fi eld yesterday, she abruptly fainted and collapsed. When her delivery suddenly seemed imminent last night in the 5th hour, we were surprised because she had not borne the child for a suffi cient number of months.
When the [village offi
cials] were informed by the fi ve-person group and swift ly assembled, a fully satisfactory ( manzoku) girl was born. We respectfully report this.
Report fi led byDen’uemon of Namashina village
15th year of Tenpō [1844], 6th month
To the Offi
ce of the Childrearing Commissioner
Confi rmed by
Th
e Headman
Th
e Head of the fi ve-man group60
Not every pregnancy went so well. In the same year and village, one Tsunohei fi led the following report:
17/04/13 3:54 PM
Subsidies and Surveillance 171
I respectfully report in writing:
In the fi rst month of this year, I reported the pregnancy of my wife, Yasu, who is 23
in this year of the dragon. However, in the 5th hour at night of the 4th day [of the 2nd month], she suddenly grew dizzy and fell from the veranda. I asked the physician Aoki Hokai to administer a medicine, but it looked as though she was going to give birth. Our neighbors and fi ve-man group reported this to the village offi cials.
In addition, the offi
cials reported this to the childrearing commissioners. [My wife]
gave birth, but because the months were too few ( tsuki tarazu yue), we were not able to keep the fetus alive and it died ( taishi). Th is I report.
15th year of Tenpō [1844], 2nd month
Namashina village
Report fi led by
Tsunohei
Group member
Shōzaemon
Witness Gizaemon
To the Offi
ce of the Childrearing Commissioner
We certify with our seals that this report is accurate.
Th
e Headman
Th
e Head of the fi ve-man group61
A few months later, one of Yasu’s neighbors fell just as she crossed a bridge on her return from a bath at her neighbor’s place. She also reported a miscarriage.62
Sudden falls appear in a large number of reports of miscarriages.63 Th is may be
because pregnant women were granted little rest from their chores, or because women calculated that a dramatic—and oft en public—fall would diff use the suspicion of foul play.
Miscarriages and stillbirths nevertheless oft en attracted close scrutiny. For example, when, in 1848, a woman in the mountains of southern Sendai miscarried aft er falling on the stones on a riverbank, the paperwork was extensive.64 Two childrearing commissioners, who also served as the headmen of two neighboring towns, collected the statements of the woman’s husband and of the head and a member of their *goningumi * or fi ve-man group. (Th ese were groups of fi ve households that shared collective responsibility for their crimes and taxes, and to which all villagers belonged.) Th
e commissioners and their witnesses fi led a detailed
forensic report on the miscarried child:
One boy
Height: 1 foot 3 inches
Both eyes: shut
Mouth shut
Right hand: clenched
Left hand: slightly extended and open
No abnormalities on penis and anus
Both legs extended
17/04/13 3:54 PM
172 Redefining Reproduction
Umbilical cord normal
No abnormalities on placenta; placed under the child’s anus Th
e entire body has no wounds and is white.
[A fetus] of about nine months
Witnessed by Childrearing Commissioners Takahashi Magojūrō and Tarōsaemon; fi ve-man group heads Miuemon and Junzō.
Th
ere followed a detailed report by the dead child’s father: My wife, who is twenty-eight years old, was pregnant with her third child. We reported on the 10th day of the 10th month of last year that she expected to deliver in the 2nd month of this year. However, on the 16th day last [of the 1st month], she gave birth to a boy as a dead fetus. We reported this immediately. I understand that as part of the offi
cial inquiry it has been asked why the child was born as a dead fetus. Twenty-three days earlier, my wife went to the riverbank to polish rice and fell on the stones. Th
ereaft er, she had sporadic abdominal pain. At dusk of the 16th day of this month, her pain grew worse. In part because we were approach-ing the month of her expected delivery, I considered this an important matter and immediately informed our relatives and group members. Both came to us straight away and wanted to fetch a doctor from the town of Shiroishi, but before they could do so, my wife gave birth to a dead fetus. Upon mature consideration, the only explanation is that when she went to the riverbank to polish rice and fell on the stones, the baby in her womb let go of her milk/breasts ( chi hanare) and was therefore born as a dead fetus. She has paid careful attention to her diet for a long time. As I have no other idea, this is all that I can report.
5th year of Kōka (1848), 1st month, 17th day
Th
e aforementioned Hanzō
Th
is report was repeated, but not quite verbatim, by a member of Hanzō’s fi ve-man group as well as two group heads and the two childrearing commissioners. All fi ve ended on the note that there was no evident reason to doubt the veracity of Hanzō’s story. All in all, this single miscarriage generated 1,600 characters of reports.
Infant deaths also required reports, investigations, and the joint signatures of childrearing commissioners, headmen, and doctors. Frequently, such reports are briefer than those of miscarriages. Th
e following example from Sendai’s northeast-
ern coast is more detailed than most in its description of the cause of death:
“Yesterday night, the baby boy suddenly ran a high fever and could not even cry, and he was unable to take milk or medicine. Th
e whole house was in upheaval, and
we fetched a doctor from [a neighboring village]. When we returned with the doctor, froth came from the baby’s mouth, and he died toward dawn today.”65
Such documents were not merely ceremonial. Childrearing commissioners frequently examined reports closely and returned them to their authors with pointed questions.66 If a birth was discovered that had not been preceded by a pregnancy
17/04/13 3:54 PM
Subsidies and Surveillance 173
report, punishments could be severe. For example, Mito imprisoned ( kingoku) and manacled men for failure to report the pregnancies of their wives, even if they resulted in healthy babies.67 Tsuyama domain issued “scoldings” to neighbors, headmen, and even landlords in cases of missing or falsifi ed reports or “careless”
infant deaths, with occasional house arrests, manacling, and degradations, such as consigning the miscreant to take the last seat in community gatherings.68 Karatsu punished even late reports, with fi ve days of labor for the child’s father and three days of labor for the other four members of his fi ve-man group.69
F O R E N S I C S TAT I S T I C S : S E C O N D - G U E S S I N G T H E
SU RV E I L L A N C E SYS T E M S
How eff ective were pregnancy surveillance systems at suppressing infanticide and abortion? Th
e statistics of most such regimes suggest a high degree of compliance.
Based on the records of more than eleven thousand pregnancies from several parts of Sendai, reported infant deaths averaged only about 12 percent.70 For a population innocent of modern medicine, this is a low rate. Th e lowest infant mortality
rates in early modern Europe and China have been observed at 9 to 15 percent.71
If we assume that every death, intentional or otherwise, was reported in Sendai, infanticides cannot have been disguised as natural infant deaths very frequently.
Reported infant deaths also fall into a plausible range in other pregnancy surveillance systems (Table 7).72
Reported stillbirths, in contrast, were extravagantly frequent in some surveillance reports. In eighteenth- and nineteenth-century populations in northern Europe, the stillbirth rate generally ranged from 1.3 to 5.6 percent.73 Th e 4.5 percent
stillbirth rate in over four thousand pregnancy reports from four districts of Sendai therefore leaves only moderate room for misreporting late-term abortions or infanticides as stillbirths. In most other surveillance regimes, ostensible stillbirth rates were far higher. In the village of Oguki in Ushiku domain, for example, eighty-seven pregnant women with at least two children claimed thirty stillbirths between them. Evidently, the rigor of surveillance varied a great deal.
Even the more thorough surveillance regimes had several possible loopholes.
Parents may simply have failed to report a pregnancy whose issue they had not yet decided. Th
at at least some couples did not report their pregnancy and were later punished is clear from the historical record. Despite its unusually harsh punishments for tardiness, Tsuyama domain found that about 2 percent of its pregnancy reports were delayed or missing.74
How many subjects of other domains ignored the reporting requirement with impunity is diffi
cult to guess. Th
ere are some tantalizing hints, however. In Sendai,
an offi
cial reported in an 1857 memorial that “due to bad weather” 120 or 130 fewer pregnancies had been registered in the north of Sendai than in the previous year,
17/04/13 3:54 PM
table 7 Pregnancy surveillance statistics Location, roughly from
Sex Stillbirths and Infant
north to south
Domain
Period
Births
ratio
miscarriages
deaths
Asai in Esashi district
Sendai
1808–1821
206
100
27 villages in Iwai district
Sendai
1808–1840
1,489
99
5.1%
19%
Nakagawa in Iwai district
Sendai
1868–1869
73
92
15%
Sanuma in Tome district
Sendai
1843
315
15%**
4 districts in Sendai (Tome,
Sendai
1843
4,330
4.5%
Monō, Motoyoshi, and
Oshika), except Sanuma
Funakoshihama in Monō
Sendai
1845–1867
238
97
1.2%**
district
Kamitozawa in Katta
Sendai
1845–1869
115
95
1.7%
10%
district
Retainers of Ichinoseki
Ichinoseki 1811–1830
515
16%**†
domain
Kozenji Ichinoseki
1811–1821
288
95
14%**
18%
Kozenji Ichinoseki
1852–1860
237
106
18%**
16%
Takabayashi Shirakawa
1861–1865
22
144
Nakaishii Hanawa
1808–1826
262
104
8–11%**††
18%
Kamiishii Hanawa
1806
16
78
31%
Otogami in Tamura
Kasama
1835–1838
19
5.0%
district
13 villages in Agatsuma
Numata
1841–1846
59
168*
21%
district
6 villages in Shimotsuke
Shogunate 1834, 1858
21
62
13%
10%
Shinchi and Oguki
Ushiku
1851–1855
102
33%**
Asō domain
Asō
1846–1850
632
114
15%**
Tsuyama castletown
Tsuyama
1782–1789,
1,223
109
4.5%
1804,
1834
Ito and Ihara
Fukuoka
1807–1808
175
12%
* Deviation from a 5 percent stillbirth rate is signifi cant at the 0.1 level.
** Deviation from a 5 percent stillbirth rate is signifi cant at the 0.05 level.
† Assumes that miscarriages were not included in the number of births.
†† Depending on whether pregnancies with unknown outcome are counted as stillbirths or live births.
note: Because epidemics such as smallpox created large fl uctuations in infant mortality from year to year, no signifi -
cance was calculated for infant deaths.
sources: Based on documents and tables in *Asō chōshi tsūshihen, * 415; *Daitō chōshi jō, * 714–719; *Esashi shishi 5, * 552–553
and 484–490; *Fujisawa chōshi honpen jō, * 512–518; *Ono chōshi shiryōhen 1 ge, * 412–415; *Ten’ei sonshi 2, * 256; Gunma kenshi *shiryōhen 12, * 793–802; *Kanuma shishi shiryōhen Kinsei 1, * 307–309; *Numata shishi shiryōhen 2, * 259–261; Tochigi kenshi *shiryōhen Kinsei 6, * 418–420; *Ushiku shishi Kinsei, * 384–385; Ohara, “Sendai-han,” 87–89, 106–111, 116–117; Kikuchi,
“Sendai-han,” 140, and “Shichigashuku,” 169–174; Sawayama, “Bushi-sō,” 86; Takagi and Mukaida, “Kinsei kokka no jinkō-gen taisaku,” 138–144; Kitō, “Tokugawa jidai nōson no nyūji shibō,” 89; Honda, Fukushima jinbutsu no rekishi 5, 98; Sawayama, *Shussan to shintai, * 135; Kalland and Pedersen, “Famine and Population in Fukuoka,” 67.
17/04/13 3:54 PM
Subsidies and Surveillance 175
and weighed the evidence for the eff ect of bad weather on the number of pregnancy reports from other years.75 Unfortunately, he did not reveal what precise mechanism he thought connected weather and pregnancies. Sendai was very far from famine conditions in 1857, and fecundity is anyway not easily reduced by any but the most severe malnutrition. Th
at couples had intercourse much more rarely just
because the weather was unpleasant is also not likely; bad weather alone should not have separated spouses on any large scale. Perhaps villagers used a moderately eff ective form of contraception. Proving that they did would be a major discovery; in the absence of further evidence to this eff ect, the only likely method is extended breastfeeding. Th
ere may also have been a surge in early-term abortions, well before the fi ft h month in which pregnancies had to be reported. Or couples may have chosen to keep off the books a pregnancy that they intended to cap with an infanticide.
Many pregnancies were reported so late (Figure 18) that we have reason to doubt the rigor of the enforcement, and perhaps also the intentions of parents who may have needed time to make up their minds about whether to raise the child.
Th
ere is actually some question as to whether early-term abortions were illegal under all pregnancy surveillance systems. While a few bans covered *ryūzan, * or
“runny births,” a term commonly applied to early-term abortions, most used the words *datai * or *ko wo orosu, * that is, “dropping” the fetus ( tai) or child ( ko) from the womb. As we saw in Chapter 3, the term *tai * was not usually employed before the fi ft h month of pregnancy. Just as in the early twenty-fi rst century, when “menstrual regulation” is a common form of pregnancy termination in countries such as Bangladesh, Edo-period advertisements for abortifacients used innocuous expressions such as “hastening the menstrual fl ow” ( gessui hayanagashi).
Infanticide continued under the pregnancy surveillance systems, if at reduced levels. In Nihonmatsu, one headman complained in a petition that “the merciless treatment of newborns has not stopped. Although several decrees have been issued, they are not strictly enforced through inspections. As a result, there is [in eff ect] no childrearing [system].”76 Heinosuke, a childrearing commissioner in Sendai, thought that surveillance simply forced people to lie. “Since the bans have become stricter, there are people who feign fetal deaths ( shitai), feign miscarriages ( hanzan), feign death by illness ( byōshi) * * a few days aft er birth, or use abortifacients ( otoshigusuri).”77 In some areas, overall fertility rates stayed suspiciously low. In the village of Oshihata in Shimōsa, the pregnancy surveillance system recorded eighteen fi rst children over a decade, twenty-four second children, seven third children, and only one fourth child.78 At the very least, surveillance failed to create large families overnight.
Th
at not every childrearing regime functioned properly is also clear from repeated reforms. One of the more suggestive was the decision of Miharu domain to ban headmen from serving as pregnancy registration offi cials ( *ninshin *
aratameyaku). While it is possible that this measure was intended to relieve the
17/04/13 3:54 PM
70
60
50
es 40
as c
ber of
30
num
20
10
0
1
2
3
4
5
6
7
8
9
10
month of pregnancy in which the pregnancy report was filed Yamakita in Tsuyama domain, 1851-1866
Kawasaki in Tsuchiura domain, 1832-1835
Namashina in Numata domain, 1844
Otogami in Kasama domain, 1835-1838
villages in Sendai, 1812, 1840, 1845-1847
figure 18. Filing time of pregnancy reports by month of pregnancy, selected villages in fi ve domains, 1812–1866. (Th
e chart assumes that pregnancies lasted ten lunar months.
sources: Sawayama, *Shussan to shintai, * 173; *Yawara no rekishi tsūshihen, * 356; Gunma *kenshi shiryōhen 12, * 795–802; *Ono chōshi shiryōhen 1 ge, * 412–415; Kikuchi, “Shichigashuku,”
169–174; *Fujisawa chōshi honpen jō, * 512–518.)
17/04/13 3:54 PM
Subsidies and Surveillance 177
headmen’s workload, the fact that it was stated as a ban suggests another motive: to end the collusion of entire village communities against surveillance. Since, unlike a headman, a pregnancy surveillance offi
cer would derive his entire authority from
his surveillance duties, he could be expected to take them more seriously.
P U N I S H M E N T S
Th
e record on punishments is small compared to the tens of thousands of surviving pregnancy reports and investigations, and the rare cases of punishment that we know about were typically light. One reason must have been the basic logic of Tokugawa law, which assessed the severity of a crime with reference to the hierarchical relationship between victim and perpetrator. A patricide could expect to die by crucifi xion, but a man who murdered his adult son would more likely face banishment.79 It was therefore diffi
cult to argue that infanticide should carry the
death penalty.
Th
ere was also a demographic argument for leniency. Nagakubo Sekisui, who in 1773 laid out in detail how a surveillance system should work, thought that a domain like his native Mito, already suff ering from depopulation, could ill aff ord to lose more subjects to banishments or executions. Death penalties for infanticide should therefore be commuted to fi nes, which Nagakubo considered both profi table and eff ective, since “infanticide originally arises from a jealous desire for money.” Th
e fi nes should be graduated according to the property of the off ender.
According to Nagakubo’s proposal, middling and smaller landholders would be fi ned several ryō, while the village elite should pay several tens of ryō for each unaccounted child.80
In 1838, the domain of Sakura announced just such a system of graduated fi nes—an act of mercy, the decree pointed out, since infanticide “deserves the death penalty.”81 Tsuyama in western Japan punished infanticides severely, but again spared the lives of the off ender. Its harshest punishment on record fell on a woman named Seki, who had her nose sliced off and was reduced to outcaste status aft er she threw her baby into a river.82 In Sendai, the harshest recorded punishments for infanticide were banishments, and even these were rare.83
Yonezawa sentenced a number of warriors, along with their neighbors, landlords, midwives, and superiors, to house arrest and dismissal from offi ce.84
Compared to contemporaneous Europe, where far fewer infants were killed outright but far more women executed for infanticide, killing babies carried little serious risk for unwilling parents even aft er the introduction of pregnancy surveillance. It is true that some people in the Tokugawa period were nonetheless put to death for infanticide, but their numbers seem to have been small and their alleged crime oft en consisted of killing a child that somebody else had entrusted to their care.85
17/04/13 3:54 PM
178 Redefining Reproduction
One of the main challenges for the designers of punishments was understanding how to break the solidarity of the village, which could easily hide abortions and infanticides. In most domains, warrior offi
cials lived in the castletown and paid
only brief, formal visits to villages, with little opportunity to inquire into the private doings of their inhabitants. Nagakubo Sekisui argued that fi nes would dampen the empathy that neighbors and village offi
cials might otherwise feel for infant killers.86
Several regimes, including that of Karatsu, held men responsible for the infanticides of their neighbors.87 Kuroha Jirōemon, whose analysis of motives we encountered in Chapter 5, agreed that any conniving neighbors should be fi ned. But if punishments were too harsh, he reasoned, people would be reluctant to inform the authorities in the fi rst place. He therefore proposed to punish the infanticidal mother by, “for example, disfi guring the hair about which women care so much.”88 Publicly infl icted shame of this kind was an offi
cial sanction at least in Shinjō, which threatened to
tattoo the foreheads of the parents, shear the mother’s hair, and shave the father’s eyebrows.89 In 1834, Tsuyama made demotion to outcaste (literally *hinin, * that is,
“nonhuman”) status the standard punishment for abortion and infanticide. Th e
announcement reasoned that even small fi nes would destroy poor commoner households; outcaste status, however, was a fi tting enough punishment, since the off enders had just demonstrated their own inhumanity.90
Perhaps in recognition of the diffi
culties of surveillance and punishment, many
domains appealed to the fear of divine retribution. Th
is was of course a major
theme in the propaganda texts and images, but a number of domains required subjects to swear oaths, oft en signed in blood, on their abstinence from abortion and infanticide. For example, subjects of Takada domain had to report to the local shrine or temple in the fi ft h month of a pregnancy and blood-print their fi ngers on the following oath:
Now that I am in the fi ft h month of my pregnancy, I shall be very careful, even with eating, and as the time of delivery approaches, even in my movements. I shall duly inform the members of my fi ve-man group and have them assemble right aft er my birth. I shall not plan any inhumane deeds. If I violate this pledge, I shall suff er the punishments of Bonten [Brahma] and Taishakuten [Indra] and all the Gods, great and small, of Japan’s sixty-some provinces, and especially of the Avatars of Izu and Hakone, the Manifest God of Mishima, the Bodhisattva Hachiman, Tenman Tenjin, and also the gods of other places. Th
is is my pledge.91
SU C C E S S F U L P O L I C I E S , P O W E R F U L SYM B O L S
In many areas, the population policies were judged a success.92 In Mito, two administrators compiled statistics that showed a dramatic increase in “births”
right aft er the introduction of subsidies.93 A poet traveling through Shimotsuke praised the eff ects of Intendant Takegaki’s infant protection measures in a waka
17/04/13 3:54 PM
Subsidies and Surveillance 179
poem.94 In Numata, the childrearing commissioners inscribed a shrine gate with the words Taigan jōju: “A great wish has been fulfi lled.”95 In Karatsu, a visiting shogunal inspector praised the success of exhortation and subsidies.96 In Sendai, an offi
cial gushed: “Th
e commoners have forgotten about the killing of newborns; the feelings of the people have come to a point that on the rare occasions when they hear that in other lands, people kill [infants], they think that these are acts like those of devils.”97
Others were openly critical of the policies. Ōhara Sakingo, a political thinker from Sendai, saw no evidence for their success.98 Satō Nobuhiro, a freelancer who favored more radical policies than mere subsidies, thought that most payments were too small to make much of a diff erence: “When people have a baby, they cannot apply themselves fully to the cultivation [of their fi elds], and their poverty increases. Can three bales of rice save them from this [predicament]? People who raise their child aft er receiving three bales of rice are not the truly poor ( gokuhin).
Th
e truly poor consider it a better strategy to have a secret abortion, and make no request for the three bales of rice.”99
Th
e doubters of the policy regimes may have had a point as far as their material eff ects were concerned. Given that the poor benefi ted most from the subsidies, why did landowners experience a greater increase in fertility in the early nineteenth century than the landless, and why was the reduction in simulated infanticides broadly similar across all wealth strata? (See Figures 6 and 7 in Chaper 5.) If there was such variance in the generosity of subsidies and rigor of surveillance, why did fertility rates rise across diff erent policy regimes?
Th
e answer may be that quite apart from promising support and threatening punishment, the policies transformed the social signifi cance of childbirth with their powerful symbolism. Even if pregnancy surveillance was a mere ritual in some villages, it was a ritual with a message. Even if subsidies fell short of off setting the cost of raising a child, the fact that a lord, a local notable, or the village community was willing to pay for other people’s children conferred a new social meaning on reproduction. Childrearing grants changed conversations about infanticide in yet another critical way: it made it that much harder for couples to claim that poverty left them no other resort, thereby silencing some of the language in which the old fertility norm had been expressed.
While the eff ect of the infanticide countermeasures is impossible to assess with precision, we can surmise that it was considerable. Map 11 plots stillbirth rates in the 1890s and shows whether infanticide countermeasures had been in force in a given district during the Edo period. In the eighteenth century, the area between Kōzuke in the south and Sendai in the north was notorious as the heartland of infanticide. Low recorded fertility (see Appendix 4) and the simulation results for Eastern Japan (see Chapter 7) show that this reputation was well deserved. If infanticide had preserved its geography from the eighteenth to the late nineteenth
17/04/13 3:54 PM
map 11. Infanticide countermeasures before 1870 and stillbirth rates in the 1890s, Eastern Japan. (District-level stillbirth fi gures do not survive for every year. Th is map is therefore
a composite of 1890 [Chiba], 1891 [Tochigi, Iwate], 1893 [Ibaraki], 1898 [Akita], 1905
[Saitama], and 1892 [all other prefectures]. source: *Meiji nenkan fuken tōkeisho shūsei. *)
17/04/13 3:54 PM
Subsidies and Surveillance 181
century, we would expect the area that on Map 11 is tiled with the squares of subsidy and surveillance programs to be gray and black with stillbirth rates as high as any in Japan, all the more so because the decades under the surveillance systems would have taught its inhabitants how to conceal infanticides as stillbirths.
Instead, we fi nd that by the 1890s stillbirths were reported most frequently at the edges of Eastern Japan. Except in the west, where the province of Echigo had long enjoyed a reputation for raising all children, the old eastern lands of infanticide were surrounded by areas with high stillbirth rates. In the southwest, 21 percent of children were reported stillborn in Suwa district. In the south, that percentage stood above 10 in all districts of Saitama prefecture, and oft en above 15. In the southeast, on the Bōsō peninsula, lay a cluster of districts in which more than a quarter of babies were reported as stillborn. In the north, too, the Iwate districts of Nakahei and Minamihei, just beyond the limits of the old policy regimes, reported very high stillbirth rates.100
Rather like the Celtic rim of the British Isles or the Buddhist periphery of South Asia, this pattern suggests the persistence of a culture at the edges of a lost core.
Subsidies and surveillance apparently did much to persuade people that infanticide was distasteful and that raising a fourth or fi ft h child was a responsible choice.
Far from everyone bought into this new worldview immediately, but the deeper in the lands of subsidies and surveillance a district lay, the fewer newborns its people killed.
Th
is spatial pattern cannot primarily have been the result of the changed economics of childrearing for two reasons. First, rather than persisting into the 1890s, such eff ects should have disappeared when subsidies and surveillance were discontinued around 1870. Second, if the direct material rather than symbolic eff ects of these subsidies can be credited with changing attitudes to childrearing, we would expect Map 11 to have a sharp perimeter retracing the extent of their sway. Yet in the 1890s, stillbirth rates were higher at the edges of the area once under subsidies and surveillance than at its center. Th
ere is no reason to believe
that the measures were less rigorously administered where a domain bordered on territories that did not combat infanticide. If subsidies and surveillance changed attitudes, those attitudes would have been spread and challenged through networks of conversations, which oft en crossed the borders of domains.
According to this hypothesis, people were likely to commit infanticide only if many of their social contacts also killed their children. Such contacts were typically spatially bound, forming a cluster around each individual. Deep inside the lands under population policy, their symbolism and material eff ects could touch every social contact of an average subject. Closer to the edges, the situation was diff erent. In southern Mito, for example, villagers were likely to fi nd spouses and interact with friends and business partners from parts of Shimōsa and Kazusa that remained innocent of infanticide countermeasures before the Meiji period. As
17/04/13 3:54 PM
182 Redefining Reproduction
a result, many people in southern Mito continued to move in social circles in which infanticide was still widely accepted. Th
ey even seem to have radiated their
tolerance of infanticide back into the central portions of Mito. If this reasoning holds, the contribution of the policy regimes is not easily distinguished from the eff ects of moral suasion. Suasion put cracks into the culture of low fertility and infanticide with words and images. Subsidies and surveillance achieved the same eff ect with force, incentives, and symbolism. Together, they tipped the numerical balance between people who endorsed infanticide and those who questioned its morality.
17/04/13 3:54 PM