Journal of Archaeological Science: Reports 34 (2020) 102627
Available online 21 October 2020
2352-409X/© 2020 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Ceramic fabrics and lead glazes of late medieval redware pots in the
Helsinki, Turku and Tallinn regions (ED-XRF, SEM-EDS)
Elisabeth Holmqvist a,*
, Tuuli Heinonen b
, Riikka V¨ais¨anen c
, Aki Pihlman d
, Andreas Koivisto c
,
Erki Russow e
a
Helsinki Collegium for Advanced Studies, P.O. Box 4, FI-00014 University of Helsinki, Finland
b
Department of Cultures, P.O. Box 59, FI-00014 University of Helsinki, Finland
c
Vantaa City Museum, Hertaksentie 1, FI-01300 Vantaa, Finland
d
Puutarhakatu 42a, FI-20100 Turku, Finland
e
Archaeological Research Collection, Tallinn University, Rüütli 10, 10130 Tallinn, Estonia
A R T I C L E I N F O
Keywords:
Redware pottery
Lead glaze
Late medieval
Finland
Estonia
ED-XRF
SEM-EDS
A B S T R A C T
We report on ceramic and glaze chemical and technological data on lead-glazed redware pottery, sampled from
late medieval, ca. 14th–17th century CE contexts at five sites across the Gulf of Finland, in modern Finland and
Estonia. Redware pots first arrived in this region as Scandinavian and Continental imports, and their local
manufacture was established in the 15th–16th centuries. We aimed to chemically characterize ceramic and glaze
recipes, distinguish between local and imported redware pots at the sites, and to examine glaze preparation and
application practices used. Altogether 34 sherds of ceramic artefacts from three hamlet sites in the Helsinki
region (Gubbacka, Mankby and Mårtensby), and the towns of Turku and Tallinn were geochemically and
microstructurally analysed by energy-dispersive X-ray fluorescence spectrometry (ED-XRF) and scanning electron
microscopy with energy-dispersive spectrometry (SEM-EDS). The majority of the analysed pots were made
of clay originating from the Gulf of Finland region, produced at least in Turku and highly likely in Tallinn.
Redware was also exchanged between these trading areas. High-lead-content (PbO < 67 wt%) glazes, sometimes
opacified with tin, were applied as lead-oxide itself or as a lead-oxide-plus-sand mixture on unfired noncalcareous,
iron-rich ceramic bodies. This glazing technology to produce impermeable pots was achievable at
relatively low temperatures and cost, hence it was commonly adopted by ceramic producers in the North, and
stayed in vogue for centuries. Redware pots from different sources appear visually and morphologically related,
yet microstructural inspection reveals varied glazing technologies.
1. Introduction and aims
In this article, we examine pottery and glaze recipes used in medieval
redware pots recovered from mainly 14th–17th century CE contexts at
archaeological sites around Helsinki, Turku and Tallinn, located in
modern Finland and Estonia, on the northern and southern coasts of the
Gulf of Finland (Fig. 1). Redware pots – often glazed, reddish-coloured,
earthenware tripod cooking pots – frequently occur in late medieval
contexts, representing early glazing technology in Northern Europe. The
lead-glazed earthenware ceramic tradition was initiated in Western
Europe (British Isles, France) between the 9th and 11th centuries CE,
spreading to the Low Countries and Germany in the course of the 12th
century (see Bliek, 1989; McCarthy and Brooks, 1988; Kilmurry, 1980).
In southern Scandinavia and on the southeast coast of the Baltic Sea,
production of these popular utensils started in the 13th and 14th centuries
– it has been suggested that the technology was brought by
specialist craftsmen migrating from continental Europe (see Davey and
Hodges, 1983; Gaimster, 1997; 1999; Elfwendahl, 1999; Holmqvist
et al., 2014). The first redware pots in the Gulf of Finland region were
imports from Southern Baltic or Scandinavian production areas, but it is
difficult to visually differentiate between separate sources due to the
stylistic and morphological similarity between products of different
manufacturers (see Christensen et al., 1994; Niukkanen, 2000; 2007;
Bergold et al., 2004; Gaimster, 2007). Written sources mention a potter’s
workshop in Tallinn as early as the 14th century and in Stockholm in
1479 (Johansson, 2007: 52; Russow, 2007: 69; Brorsson, 2016), without
* Corresponding author.
E-mail address: elisabeth.holmqvist@helsinki.fi (E. Holmqvist).
Contents lists available at ScienceDirect
Journal of Archaeological Science: Reports
journal homepage: www.elsevier.com/locate/jasrep
https://doi.org/10.1016/j.jasrep.2020.102627
Received 20 May 2020; Received in revised form 11 September 2020; Accepted 29 September 2020
Journal of Archaeological Science: Reports 34 (2020) 102627
2
mentioning what kind of pottery was made. From the archaeological
sources, clear growth in the use of redware in Tallinn is visible from the
second half of the 15th century (Russow et al., 2019: 196), although the
direct evidence of medieval local production of redware (kilns, tools,
wasters, etc.) is still lacking. In Finland, there is a notable increase in
redware sherds encountered in the 15th century contexts in the town of
Turku, and this has been interpreted as a possible marker of locally
established manufacture of redware pottery (Pihlman, 1989: 104).
However, the earliest direct evidence of redware manufacture in Turku
dates only to the turn of the 16th and 17th centuries (Tulkki, 2003).
Our aim was to identify different ceramic and glaze recipes, distinguish
between local and imported redware pots at the sampled sites, and
to examine glaze preparation and application practices used in redware
manufacture. Altogether 34 sherds of ceramic artefacts (Table 1, Figs. 2
and 3) were sampled from excavation materials from three hamlet sites
(Gubbacka, Mankby and Mårtensby) located in the modern greater
Helsinki area on the northern coast of the Gulf of Finland, and the towns
of Turku and Tallinn, in modern coastal Finland and Estonia, respectively
(Fig. 1). The ceramic samples were prepared as polished crosssections
and subjected to geochemical and microstructural characterization
via energy-dispersive X-ray fluorescence spectrometry (ED-XRF)
and scanning electron with energy-dispersive spectrometry (SEM-EDS)
at University of Helsinki laboratories. The samples were assigned to
fabric groups indicated by statistical processing of the geochemical EDXRF
data, confirmed by ceramic paste and mineralogical compositions
acquired by SEM-EDS. The glaze composition, preparation and application
methods were analysed using SEM-EDS imaging and microchemical
tools.
2. Archaeological sites and ceramic artefacts
The medieval hamlet site of Gubbacka is located on a hill in the
modern port of Helsinki (the Vuosaari district), with direct access to the
sea (see the site locations in Fig. 1). Gubbacka was established by
Swedish settlers in the late 13th century and deserted by the late 16th
century. The Gubbacka excavations in 2002–2003 and 2008–2010
produced a large assemblage of redware pottery finds, dating mainly
from the late 14th to the 16th centuries CE (Koivisto et al., 2010;
Holmqvist et al., 2014; see Fig. 2 for sample nos. 1–10).
In general, most of the pots sampled for this project represent
different ceramic morphologies and typo-chronologies, and are often
fairly small-sized rim or body sherds. Thus, it is not always possible to
determine the vessel morphology with certainty. Most of the ceramic
sherds from Gubbacka have reddish or reddish-greyish fabrics and
glazed interiors (see Table 1 for sample descriptions). Two samples (nos.
3 and 7) are unglazed greyish coarseware sherds, the former dated to the
14th century and the latter, of “Iron Age” type fabric, presumably local
and TL dated to the 15th century (see Holmqvist et al., 2014). Certain
glazed sherds (nos. 4 and 8–10) were selected as suspected imports
based on their visual appearance and fabric quality. In addition to the
artefacts, fragments of burnt clay daubs (nos. 35–36) were sampled from
two furnaces at the Gubbacka site (dating to the 15th and 14th–17th
centuries, respectively) as likely representatives of the local clay
composition.
Mankby is located on a wooded hill in western Espoo (west of Helsinki),
along the River Mankki close to the Espoo bay. Mankby was
founded by Swedish settlers by the early 13th century, and the village
site was abandoned in the 1550s. The site was excavated in 2007–2013,
and the rich find material included redwares dating to the 15th and 16th
centuries. The large number of imported objects found suggests that the
people living in Mankby were actively involved in trade (Harjula et al.,
2016). Five pots (Fig. 2, Table 1, nos. 11–15) were selected from the
assemblage at Mankby. Glazed sherd no. 15 has an unusual, overlapping
rim profile and a yellowish glaze. It is uncertain whether the handle
fragments (nos. 13–14) belonged to glazed or unglazed pots.
The urban settlement of Tallinn on the shore of Tallinn Bay was
established some time after the Christianization of Northern Estonia in
1219/1220 and acted as a busy international merchant town from the
mid-13th century at latest (Russow, 2016). From the origins of the town
until the early modern period, its population was very diverse and
attracted migration from Scandinavia, coastal Finland, and northern
Germany (Johansen, 1951). The redware sherds (nos. 16–20, no. 18
being a glazed floor tile; Table 1) sampled from Tallinn date to the late
14th century – first half of 15th century CE, and were found at a suburban
site that was situated not far from the river crossing in the medieval
period. The location and the character of the main structure, and
Fig. 1. Map of the sampled sites (circles) and locations mentioned in the text.
E. Holmqvist et al.
JournalofArchaeologicalScience:Reports34(2020)102627
3
Table 2
ED-XRF results of ceramic fabrics (mean values of 3–5 measurements per sample). Results normalized to 100%, sample order and group allocations as indicated by the CA dendrogram.
No Date Glaze Site Na2O
(%)
MgO
(%)
Al2O3
(%)
SiO2
(%)
P2O5
(%)
SO3
(%)
K2O
(%)
CaO
(%)
TiO2
(%)
MnO
(%)
Fe2O3
(%)
SiO2 /Al2O3 NiO (ppm) CuO (ppm) ZnO (ppm) Rb2O (ppm) SrO (ppm) Y2O3 (ppm) ZrO2 (ppm) BaO (ppm)
Fabric 1a
33 1550–1650 x Mårtensby 1.48 1.70 15.50 71.10 0.90 0.18 2.69 0.94 0.62 0.03 5.40 4.59 14 2 86 109 87 45 363 264
34 1550–1650 x Mårtensby 1.61 1.79 15.60 70.50 0.82 0.20 2.67 0.97 0.61 0.03 5.52 4.52 1 116 111 83 33 363 295
12 c. 1450–1550 – Mankby 1.38 1.96 16.90 68.90 0.75 0.20 3.22 0.57 0.61 0.03 6.28 4.08 20 74 86 129 78 27 273 358
7 15th c. – Gubbacka 1.97 1.84 18.00 66.50 0.57 0.29 3.00 1.08 0.69 0.03 6.43 3.69 42 106 136 106 35 273 528
35 15th–16th c. – Gubbacka daub 4.45 1.78 13.60 67.00 0.70 0.28 3.62 2.18 0.61 0.04 5.29 4.93 110 154 268 23 395 599
36 15th–16th c. – Gubbacka daub 3.41 1.91 14.80 65.50 1.65 0.29 3.26 2.05 0.62 0.05 5.95 4.43 3 163 129 264 24 287 634
15 16th c.? x Mankby 3.31 2.22 15.60 65.30 1.01 0.19 3.65 1.48 0.58 0.06 6.38 4.19 112 156 133 38 255 471
27 1750–1800 x Turku 3.77 2.23 15.40 66.60 0.50 0.26 3.23 0.97 0.55 0.03 6.00 4.32 78 89 152 133 35 298 562
13 c. 1450–1550 – Mankby 2.73 2.58 15.00 67.80 0.72 0.19 3.19 1.18 0.59 0.05 5.77 4.52 15 25 84 116 134 35 259 421
11 c. 1450–1550 x Mankby 3.46 2.18 15.60 67.20 0.76 0.16 2.79 1.29 0.50 0.06 5.72 4.31 51 10 92 149 116 33 252 473
28 1750–1800 – Turku 3.68 2.65 15.30 66.00 0.35 0.24 3.02 1.11 0.52 0.06 6.76 4.31 1 22 89 150 120 30 204 458
3 14th c. – Gubbacka 2.70 1.81 18.10 61.70 3.59 0.27 2.82 1.03 0.60 0.02 7.07 3.41 43 93 138 160 30 250 593
22 late 13th c. – Tallinn 3.12 2.41 17.40 61.70 1.55 0.21 3.43 1.94 0.52 0.07 7.43 3.55 29 40 100 137 168 34 266 485
μ (n ¼ 13) 2.85 2.08 15.91 66.60 1.07 0.23 3.12 1.29 0.59 0.04 6.15 4.22 22 31 102 136 142 32 288 472
σ 1.0 0.3 1.3 2.8 0.8 0.0 0.3 0.5 0.1 0.0 0.7 0.4 17.0 27.4 21.3 16.3 61.3 5.9 54.4 115.0
Fabric 1b
29 Late 16th–early 17th c. – Turku 3.53 3.37 17.40 61.90 0.41 0.18 3.29 1.33 0.68 0.06 7.48 3.56 33 136 162 124 34 241 481
30 Late 16th–early 17th c. – Turku 3.68 3.33 17.00 62.50 0.41 0.17 3.31 1.34 0.64 0.05 7.32 3.68 44 27 114 155 125 34 229 493
31 17th c.? x Mårtensby 3.69 3.88 18.90 59.10 0.70 0.09 3.43 1.32 0.69 0.05 7.74 3.13 115 92 149 145 31 182 550
32 Iron Age? – Mårtensby 3.60 4.01 18.80 55.40 2.36 0.26 3.16 1.75 0.82 0.06 9.54 2.95 34 68 178 170 140 19 185 444
μ (n ¼ 4) 3.63 3.65 18.03 59.73 0.97 0.17 3.30 1.44 0.71 0.06 8.02 3.33 39 61 130 159 134 30 209 492
σ 0.1 0.3 1.0 3.2 0.9 0.1 0.1 0.2 0.1 0.0 1.0 0.3 7.1 40.4 36.7 9.1 10.6 7.1 30.2 43.9
Fabric 2a
8 Early 16th c. x Gubbacka 1.61 1.07 10.50 77.70 1.16 0.22 2.12 0.56 0.69 0.03 5.07 7.40 13 76 88 102 26 401 324
10 Early 16th c. x Gubbacka 0.98 10.90 77.90 0.79 0.21 2.67 0.75 0.57 0.03 4.80 7.15 36 72 97 103 41 428 417
19 Late 14th–early 15th c. x Tallinn 1.60 1.54 12.30 74.60 0.48 0.19 2.57 1.22 0.59 0.05 5.03 6.07 19 53 95 79 34 356 350
26 1550–1650 x Turku 1.63 1.50 13.10 75.00 0.44 0.23 2.35 0.87 0.59 0.04 4.60 5.73 31 95 116 93 59 389 279
16 Late 14th–early 15th c. x Tallinn 1.66 1.52 12.90 73.60 0.73 0.26 2.59 1.64 0.58 0.02 4.57 5.71 22 75 97 104 85 26 287 351
24 Late 15th c. x Turku 2.22 1.43 11.60 74.20 0.77 0.40 2.57 1.33 0.59 0.06 4.96 6.40 5 40 88 94 120 51 419 399
23 Late 15th c. x Turku 2.01 1.34 12.70 74.80 0.32 0.33 2.78 0.78 0.53 0.03 4.60 5.89 20 155 59 99 85 40 336 385
18 Late 14th–early 15th c. x Tallinn 1.82 1.73 12.40 72.20 1.18 0.44 1.81 3.62 0.54 0.06 3.81 5.82 17 35 123 103 149 41 292 404
6 c. 1450–1600 x Gubbacka 1.92 1.52 13.70 70.90 0.65 0.04 2.53 1.04 0.53 0.02 5.70 5.18 16 49 65 112 84 3 330 239
μ (n ¼ 9) 1.81 1.40 12.23 74.54 0.72 0.26 2.44 1.31 0.58 0.04 4.79 6.15 16 50 81 101 100 36 360 350
σ 0.2 0.2 1.0 2.3 0.3 0.1 0.3 0.9 0.0 0.0 0.5 0.7 6.6 43.1 22.1 8.9 22.4 16.2 52.5 60.3
Fabric 2b
1 16th c. x Gubbacka 1.56 14.70 71.90 0.62 0.20 3.13 0.76 0.75 0.02 5.95 4.89 21 77 130 101 42 424 433
4 16th c. x Gubbacka 1.66 1.47 14.50 71.10 1.30 0.22 2.75 0.71 0.78 0.02 6.04 4.90 15 16 99 110 89 41 425 611
2 c. 1450–1600 x Gubbacka 1.78 1.37 14.20 71.10 0.60 0.30 3.49 1.06 0.69 0.03 6.19 5.01 42 90 124 110 46 385 419
14 c. 1450–1550 ? Mankby 2.89 2.10 14.00 68.70 0.84 0.17 3.09 1.27 0.66 0.05 5.93 4.91 28 111 126 144 30 351 399
17 Late 14th–early 15th c. x Tallinn 1.86 1.96 14.30 70.50 0.65 0.29 2.69 1.87 0.68 0.05 6.21 4.93 89 11 92 109 126 40 322 341
9 15th c. x Gubbacka 1.72 1.57 13.50 69.70 2.65 0.23 2.74 1.35 0.63 0.02 6.05 5.16 9 109 101 160 42 362 601
20 Late 14th–early 15th c. x Tallinn 1.76 1.64 12.50 73.00 0.56 0.24 2.84 1.30 0.60 0.04 5.65 5.84 2 157 74 117 106 42 282 367
25 16th c. x Turku 1.81 1.55 12.20 70.90 1.68 0.44 2.24 4.20 0.63 0.04 4.46 5.81 4 14 58 99 124 50 312 265
5 c. 1450–1600 x Gubbacka 1.92 1.46 12.60 74.10 0.51 0.18 2.67 1.09 0.70 0.13 5.35 5.88 15 32 97 112 88 48 288 398
μ (n ¼ 9) 1.93 1.63 13.61 71.22 1.05 0.25 2.85 1.51 0.68 0.04 5.76 5.26 22 40 90 114 116 42 350 426
σ 0.4 0.2 0.9 1.6 0.7 0.1 0.4 1.1 0.1 0.0 0.6 0.4 33.1 48.3 17.3 10.9 24.4 5.7 53.8 113.5
Fabric 3
21 Late 14th–early 15th c. x Tallinn 2.93 2.49 16.40 49.82 0.48 1.35 2.94 0.90 0.43 0.03 13.46 3.04 36 24 134 823 729 352 1568 3451
E.Holmqvistetal.
Journal of Archaeological Science: Reports 34 (2020) 102627
4
the variety of the artefacts, suggest that a significant proportion of the
late medieval finds, including the samples, are connected to an inn or
some other public institution (Kadakas et al., 2013; Russow et al., 2013).
The ceramic fragment from Sulevim¨agi 4/6 (no. 21) comes from a
plot beside the town wall that was used by the northwest Russian traders
and craftsmen during the 13th and 14th centuries (Lightfoot et al., 2016:
85). This sherd was initially interpreted as production waste of glazed
redware but subsequent research revealed that it is a cupel used for
making small seed beads. Similar cupels as well as semi-finished products
have been found in the Russian quarters in Riga, Latvia and Tartu,
Estonia, making it plausible that the Tallinn finds are part of the craft
activities of Novgorodian and Pskovian artisans (see Caune, 2004;
Russow, 2019). The final Tallinn sample (no. 22) was taken from an
unglazed, reddish coarse-ware tripod pot, found almost intact at the
corner of a medieval basement at number 5 Apteegi Street, close to the
market square and town hall. From the technological and typological
standpoint, this pot represents rather typical 13th–14th century kitchenware,
used in bulk everywhere around the town, although more
sizable and globular in shape than usual. Stylistically, these pots (for
basic typology, see M¨all and Russow, 2004: Fig. 5C–E) resemble tripods
used in the Roskilde-Lund region in southern Scandinavia (see published
in Bencard and Roesdahl, 1972: Cat. no. 29–30; Wahl¨o¨o, 1976: Cat. no.
182–184; Johansson, 2013), which might indicate that potters migrated
from there when northern Estonia including Tallinn was under the
Danish rule (that is, until the mid-14th century). The Tallinn pots are of
reddish fabric, and glazed ones (nos. 16, 19, 20) show dark reddish
surfaces, excluding pot no. 17 which is dark greyish in colour (with a
typical redware rim form), and the floor tile (no. 18) which has a
Table 1
List of the analysed ceramic artefacts with context information.
No Context / Catalogue
ID
Site Description Glazed Suggested Date
1 KM2008043:49 Gubbacka Pot rim, reddish, fine-grained ware, wheel-made Interior 16th c.
2 KM2008043:86 Gubbacka Pot (?) bodysherd, very soft, coarse material, light reddish ware Interior c. 1450–1600 (context)
3 KM2008043:94 Gubbacka Pot rim, reddish-brown coarse ware Unglazed 14th c.
4 KM2009083:122 Gubbacka Pot? Bodysherd, yellow–brown coarse ware Interior 16th c.
5 KM2009083:138 Gubbacka Pot bodysherd, reddish-bown ware, hard fired Interior c. 1450–1600 (context)
6 KM2010077:26 Gubbacka Pot bodysherd, reddish ware Both surfaces c. 1450–1600 (context)
7 KM2010077:181 Gubbacka Bodysherd, coarse grey earthernware Unglazed 15th c. (TL dated)
8 KM2010077:233 Gubbacka Pot bodysherd, light reddish relatively fine grained ware Interior Early 16th c.
9 KM2010077:240 Gubbacka Pot bodysherd, yellowish/reddish fine grained ware Both surfaces 15th c.
10 KM2010077:255 Gubbacka Pot bodysherd, wheel-made, ligh reddish, relatively fine-grained
ware
Interior, glaze spots
on exterior
Early 16th c.
11 KM2008044:277 Mankby Pot bodysherd, reddish/brownish ware, hard fired Interior c. 1450–1550 (context)
12 KM2008044:376 Mankby Pot bodysherd, reddish fine-grained ware Unglazed c. 1450–1550 (context)
13 KM2008044:401 Mankby Pot handle, reddish ware Unglazed handle c. 1450–1550 (context)
14 KM2009032:35 Mankby Pot handle, light reddish ware Unglazed handle c. 1450–1550 (context)
15 KM2011014:187 Mankby Pot rim, overlapping rim, light reddish/brownish ware Interior, glaze spots
on exterior
16th c.?
16 AI7032:1557 Tallinn Pot, reddish/brownish ware Interior and rim End of the 14th–first half
of the 15th c.
17 AI7032:1623 Tallinn Pot, dark greyish fine-grained ware, wheel-made, hard fired Interior and rim End of the 14th–first half
of the 15th c.
18 AI7032:1623 Tallinn Floor tile, dark reddish coarse ware Exterior End of the 14th–first half
of the 15th c.
19 AI7032:1625 Tallinn Pot, reddish fine-grained ware Interior and rim End of the 14th–first half
of the 15th c.
20 AI7032:1175 Tallinn Pot (?) bodysherd, light reddish ware, wheel-made Interior End of the 14th–first half
of the 15th c.
21 AI6648:8 Tallinn Sulevim¨agi 4/6 Ceramic cupel, coarse brown/red ware Coating (glaze?) on
both surfaces
22 AI6713: 270 Tallinn Apteegi street Tripod pot rim, coarse grained, light reddish ware, grey core,
wheel-made
Unglazed Late 13th c.
23 TMM 22,367
KE1034:006
Turku
Tuomiokirkontori
Pot rim, orange fine-grained ware, hard-fired Interior Later half of the 15th c.
24 TMM22367
KE2154:001
Turku
Tuomiokirkontori
Tripod pot base, fine grained, hard-fired Interior, glaze spots
on exterior
Later half of the 15th c.
25 TMM22196
KE548:003
Turku Rettinginrinne Tripod pot base, brownish/reddish fine-grained ware, hard-fired Interior, glaze spots
on exterior
16th c.
26 TMM22196
KE551:004
Turku Rettinginrinne Pot rim, fine light-reddish ware, thin-walled, hard-fired Interior 1550–1650
27 TMM22890
KE153:010
Turku Linnankatu 35b Pot bodysherd, light reddish, medium-fine ware, finger
impressions, two rings on handles, hard-fired, kiln waste
Glaze spots on both
surfaces
1750–1800
28 TMM22890
KE153:052
Turku Linnankatu 35b Bodysherd, decorative grooves on exterior, wheel-made, finegrained
light reddish ware, hard-fired, kiln waste
Unglazed 1750–1800
29 TMM18335:259 Turku
Tuomiokirkkokatu 2–4
Pot rim with broken handle, finger-impressions, wheel-made, fine
reddish ware, thin-walled, hard-fired, kiln waste
Unglazed Late 16th–early 17th c.
30 TMM18335:369 Turku
Tuomiokirkkokatu 2–4
Pot rim with handle, finger impressions, wheel-made, light
reddish, relatively fine-grained ware, hard-fired, kiln waste
Unglazed Late 16th–early 17th c.
31 KM2011018:1026 Mårtensby-11 Alue 4 Pot bodysherd and handle Interior 17th c.?
32 KM39163:686 Mårtensby-12 Alue 4 Grey earthenware bodysherd Unglazed Iron Age? Secondary
context
33 KM39466:305 Mårtensby-13 Alue 3 Pot bodysherd Both surfaces? 1550–1650
34 KM39466:317 M¨artensby-13 Alue 3 Pot rim Glazed 1550–1650
35 KM2010077:612
R401
Gubbacka Clay daub fragment from furnace Unglazed 15th c.
36 KM2010077:679
R601
Gubbacka Clay daub fragment from furnace Unglazed 14th–17th c.
E. Holmqvist et al.
Journal of Archaeological Science: Reports 34 (2020) 102627
5
Fig. 2. Ceramic samples from Gubbacka (nos. 1–10), Mankby (nos. 11–15), and Tallinn (nos. 16–22).
E. Holmqvist et al.
Journal of Archaeological Science: Reports 34 (2020) 102627
6
Fig. 3. Ceramic samples from Turku (nos. 23–30), and Mårtensby (nos. 31–34).
E. Holmqvist et al.
Journal of Archaeological Science: Reports 34 (2020) 102627
7
yellowish glaze.
Eight ceramic samples (nos. 23–30), including rim and handle fragments,
were selected from four excavation sites by the Museum Centre of
Turku, from contexts dating from the late 15th century until 1800 CE.
The medieval town of Turku is located on the southwest coast of Finland
on the River Aura (Fig. 1). At the river mouth near the town, there was a
Crown castle and the inland roads met in the town. The settlement in the
Aura river valley was dense by the end of the Iron Age (ca. 1150/1200
CE). In the 13th century it became the first ecclesiastical centre of
Finland: the bishop’s residence was built 1.5 km upstream on the cape of
Koroinen and around 1300 CE, Turku was founded (Ratilainen et al.,
2016; Harjula et al., 2018). In the Middle Ages and long afterwards,
Turku was the most prominent town in the eastern province of the
Kingdom of Sweden.
Three of the Turku sites are located in the area of the medieval town
(Tuomiokirkontori, Rettiginrinne and Tuomiokirkkokatu 2–4), and one
in the area inhabited only in post-medieval times (Linnankatu 35b). The
Tuomiokirkontori (Cathedral Square) excavations in 2005–2006 were
located in the centre of the medieval town, near the cathedral on its
southeast, south and southwest sides. The oldest urban deposits
discovered date to the early 14th century (Pihlman, 2010). Sherd nos. 23
and 24 are from Tuomiokirkontori deposits dating to the second half of
the 15th century, found from the south/southeast side of the cathedral.
The Rettiginrinne (Rettig Rise) excavations in 2000–2001 were carried
out on the edge of the medieval Convent quarter, under Vartiovuori
(Guard Hill), in the area built in the second half of the 14th century
(Pihlman, 2010: 19–21; Ratilainen, 2010: 43-44; Saloranta, 2010: 65).
Sherd no. 25 from this site was discovered in a 16th-century deposit,
while no. 26 is from a deposit dated between ca. 1550–1650.
The excavations at the plot at number 35b Linnankatu (Castle Street)
were executed in 2012, on the western side of the River Aura, ca. 1.5 km
from the cathedral towards the river mouth, on the edge of the town
built in the second half of the 17th century. Written sources revealed
that pot makers had lived there for five generations from the 1680s until
1832. The analysed fragments (nos. 27 and 28) were from unfinished
vessels (kiln waste). Their context dates to the second half of the 18th
century and was rich in fragments of stove tiles and pottery broken
during the production process (Pihlman and Savolainen, 2019).
The excavations on the northern edge of the medieval town on the
Tuomiokirkko (Cathedral) Street were conducted in 1976–1977. Sherd
nos. 29 and 30 are unfinished objects derived from a deposit containing
a large amount of stove tile and pottery production waste, the waste
dating from the turn of the 16th and 17th centuries (Tulkki, 2003). Of
the Turku pots, nos. 23–27 have greyish, glazed surfaces, whereas nos.
28–30 are unglazed with a light reddish fabric.
Furthermore, four sherds (nos. 31–34; Fig. 3) dated to the turn of the
16th and 17th centuries were sampled from the village of Mårtensby,
two of which (nos. 31 and 33) had dark greyish glazed surfaces. Mårtensby
is in the countryside around the modern-day town of Vantaa, on
the shore of River Vantaa some 10 km from the seashore. The village was
founded by Swedish immigrants in the 13th century. Excavations at one
of the farm sites, Lillas, were carried out in 2011–2013, and a large
assemblage of redware pottery was collected, dating mainly between the
16th and 18th centuries (Kadakas and V¨ais¨anen, 2012; Koivisto, 2014;
Heinonen, 2015; V¨ais¨anen, 2016). The inhabitants of Lillas were
engaged in large-scale peasant trade in the early 16th century and had
lively contacts to Tallinn. In those days, Tallinn was the town nearest to
the villages in what is now the Helsinki region, situated only 90 km
south across the Gulf of Finland (one day’s journey) in modern Estonia.
Active contacts across the Gulf of Finland are recorded in written sources
between the 14th and 16th centuries, and it has been suggested that
redwares were first brought to Finland from or via Tallinn (Salminen,
2012; Holmqvist et al., 2014).
3. Methods
3.1. Sample preparation
Redware sherds in archaeological contexts especially in Finland are
often small-sized and thus do not allow extensive invasive sampling to
produce a homogeneous sample for quantitative analysis (e.g. minimum
of 5 g of material for pressed pellet preparation, see Holmqvist, 2017 and
references therein). Furthermore, cross-section samples of ceramics
facilitate the analysis of different phases, e.g. the glaze coatings, glazefabric-interfaces
and mineralogy, and it is also possible to focus the
paste/fabric analysis on the fabric cores, which are less sensitive to postFig.
4. Ceramic cross-sections of samples in polished resin blocks: A = lead-glazed no. 1 from Gubbacka; B = unglazed no. 7 from Gubbacka, C = tin–lead-glazed no.
15 from Mankby; D = lead-glazed no. 17 from Tallinn.
E. Holmqvist et al.
Journal of Archaeological Science: Reports 34 (2020) 102627
8
depositional or coating-absorbed elemental contamination than nearsurface
areas. Therefore, we chose a sampling strategy which maximized
sample preservation (a profile cut of ca. 1 cm) and the number of
applicable analytical methods (visual and microscopic examination of
fabrics, ED-XRF spot analysis, SEM-EDS microstructural and microchemical
analysis) as well as allowing us to control the lead enrichment
in the ceramic data (by avoiding areas near the coated surfaces in paste/
fabric analysis). In addition, chlorine values, possibly enriched by the
resin material especially in the case of porous ceramics, and P2O5 values
prone to post-depositional variation, were excluded from the statistical
tests.
The analytical specimens were cut with a Buehler diamond saw
perpendicular to the vessel surface and mounted in resin blocks (Fig. 4),
allowing sample longevity and repeated analysis. The cross-sections
were then polished with diamond paste (down to a grain size of 0.5
µm) and carbon coated for the SEM-EDS analysis to eliminate charging
effects.
3.2. ED-XRF analysis of ceramic cross-sections
The instrument utilized for the ED-XRF analysis was a Rigaku NEXDE
VS bench top ED-XRF spectrometer based in the Laboratory of
Archaeology at the University of Helsinki. The instrument was operated
in point analysis mode with the beam diameter adjusted to 1 mm to
analyse major, minor and trace elemental concentrations of the ceramic
fabrics. The reported results are normalized mean values of 3–5
measured points, selected carefully avoiding large inclusions and areas
close to the vessel surfaces to avoid mineralogical effects and surface
Fig. 5. Cluster analysis dendrogram of the ED-XRF ceramic fabric data showing two main clusters with subgroups 1a–b and 2–b.
E. Holmqvist et al.
Journal of Archaeological Science: Reports 34 (2020) 102627
9
contamination, particularly absorption of Pb from the glaze coatings.
The data was acquired in a helium atmosphere and using a tube voltage
of 60 kV, 35 kV, 6.5 kV and a measuring time of 60, 60 and 100 s for
high-Z, mid-Z and low-Z elements, respectively. The results were
quantified using the spectrometer’s software and fundamental parameters
(see Supplementary Table 1 for the complete ED-XRF data set).
The data quality was controlled by analysing a standard reference
sample, NIST 76a (see Supplementary Table 2). The fabric data of the
glazed artefacts shows increased Pb values deriving from the coatings
(the most extreme effect is seen on nos. 6 and 21), yet unaffecting the
data patterns when normalized with or without Pb. The precision and
accuracy tests (Supplementary Table 2) on the standard reference material,
Burnt Refractory NIST 76a, show good precision values with
relative variation coefficients being below 1.5% for all the oxides with
concentration values above 0.3 wt%. For the accuracy tests, the average
results compared to the certified value show relative errors lower than
10% for all reported oxides with contents above 0.5 wt%.
The IBM SPSS 25 software was used for statistical data processing.
The fabric groups of the samples were formed based on the cluster
analysis (CA) groupings of the ED-XRF data using Ward’s Squared
Euclidian method (see cluster analysis dendrogram in Fig. 5), and the
data patterns were further tested using principal component analysis
(PCA, Fig. 6a–b). The ED-XRF measured concentrations of MgO, Al2O3,
SiO2, TiO2, Fe2O3, ZrO2, Rb2O and SrO (Table 2) were used in both
statistical approaches. The groups indicated by the ED-XRF data were
further verified with the mineralogical and ceramic paste compositional
data acquired by SEM-EDS.
3.3. SEM-EDS analysis of ceramic bodies, mineralogy and glaze coatings
The instrument applied in the SEM-EDS data acquisition was a
Hitachi S-4800 high-resolution field emission scanning electron microscope
(FE-SEM) based in the Laboratory of Inorganic Chemistry at the
University of Helsinki. The samples were examined using both backscatter
(BSE) and secondary electron (SE) imaging of the ceramic crosssections
in order to observe the ceramic microstructure, grain size,
surface treatment, and mineral composition. These features were
documented by micrographs taken with different magnifications. Prior
Fig. 6. Principal component analysis plots of ED-XRF ceramic fabric data, a) samples marked by CA groupings (the first three components represent 94.43% of the
total variation); b) the component plot.
E. Holmqvist et al.
Journal of Archaeological Science: Reports 34 (2020) 102627
10
to the SEM analyses, the ceramic cross-sections were examined under a
stereomicroscope (Leica M80, Fig. 4) to facilitate visual identification of
surface layers, inclusions and voids in the fabrics. For elemental analysis
of ceramic pastes, mineral grains and glazes, the SEM was equipped with
an Oxford Instruments 350 INCA energy-dispersive X-ray microanalysis
system (SEM-EDS) under the following conditions: working distance 15
mm; accelerating voltage 20 kV; process time 5, equivalent of detector
deadtime ca. 30%; time of acquisition 180 s.
The reported elemental concentrations of ceramic pastes (Table 3)
are mean values of 3–5 SEM-EDS measurements of 250 × 250 µm sized
areas (equivalent to an image of the body area at 500 × magnification),
selected by avoiding large mineral particles that were probed separately.
The reported elements for the ceramic pastes are Na2O, MgO, Al2O3,
SiO2, P2O3, K2O, CaO, TiO2 and FeO; other concentrations below 0.5 wt
% were below the detection limit of the SEM-EDS instrument. The
measurements were checked for consistency, recalculated by stoichiometry
as oxides using Oxford INCA software and reported as average
weight percent values of oxides. For the glaze compositions (Table 4),
3–5 results were acquired by measuring areas of varying sizes, avoiding
unfused grains, glaze surface and glaze-body interface areas. The data
precision and accuracy tests from the time of the SEM-EDS analyses are
reported in Holmqvist et al. 2018 (relative variation co-efficients in
precision tests < 10% or < 20%; relative errors in accuracy tests < 10%
or < 20% for all oxides). Individual mineral grains in glazes and ceramic
bodies were probed separately, and the grain size and frequency, glaze
thickness and microstructure, and the estimation of the vitrification
stage of both glazes and ceramic bodies as indicators of firing temperatures
were evaluated in microstructural analyses (see Holmqvist, 2019:
105–107 and references therein).
4. Results and discussion
4.1. Ceramic fabric groups based on ED-XRF and SEM-EDS data
The cluster analysis dendrogram of the ED-XRF ceramic data (Fig. 5)
shows two main branches and their subgroups (Fabrics 1a–b and 2a–b in
Fig. 5, Table 2). Similar data patterns are shown in the PCA plots of the
ED-XRF data (Fig. 6a–b). Fabric group 1 is dominated by artefacts (11
Table 3
SEM-EDS data of the ceramic pastes. All oxides normalized to 100%. Reported values are means of 3–5 measurements expressed as wt% of oxides (by stoichiometry).
Sample order and group allocations as indicated by the CA dendrogram.
No Site Na2O (%) MgO (%) Al2O3 (%) SiO2 (%) SiO2 / Al2O3 P2O5 (%) K2O (%) CaO (%) TiO2 (%) FeO (%)
Fabric 1a
33 Mårtensby 0.76 1.93 17.63 62.76 3.56 0.81 3.99 1.11 1.00 10.01
34 Mårtensby 0.85 1.98 17.56 65.05 3.70 0.79 4.06 1.10 0.92 7.69
12 Mankby 1.20 2.04 18.80 63.31 3.37 0.50 4.30 1.02 1.02 7.82
7 Gubbacka 1.04 1.79 18.94 62.90 3.32 1.53 4.03 1.03 1.18 7.56
35 Gubbacka daubt 2.47 1.95 13.88 69.62 5.02 0.64 3.06 1.88 0.88 5.63
36 Gubbacka daubt 1.76 2.32 15.82 65.59 4.15 1.17 3.96 1.68 0.90 6.79
15 Mankby 1.49 2.40 15.79 63.34 4.01 0.80 5.08 1.30 0.92 8.88
27 Turku 1.65 2.79 17.47 63.14 3.61 0.39 4.51 1.12 0.87 8.07
13 Mankby 1.38 3.31 17.55 64.30 3.66 0.70 3.99 1.27 0.82 6.67
11 Mankby 1.60 2.34 17.05 62.87 3.69 0.95 4.82 1.96 0.88 7.55
28 Turku 1.71 2.99 17.28 62.98 3.64 0.37 4.40 1.07 0.85 8.36
3 Gubbacka 0.93 2.04 17.70 59.84 3.38 3.22 4.24 1.18 1.09 9.76
22 Tallinn 0.93 2.63 19.15 60.86 3.18 0.85 4.52 1.87 0.99 8.20
μ (n ¼ 13) 1.37 2.35 17.28 63.58 3.72 0.98 4.23 1.35 0.95 7.92
σ 0.48 0.46 1.44 2.36 0.47 0.74 0.49 0.36 0.10 1.20
Fabric 1b
29 Turku 1.66 3.36 18.50 59.98 3.24 0.22 4.91 1.49 1.06 8.83
30 Turku 1.67 3.57 18.71 59.54 3.18 0.33 4.94 1.30 0.95 8.99
31 Mårtensby 1.58 3.41 19.18 58.74 3.06 0.66 4.93 1.39 1.04 9.06
32 Mårtensby 1.58 3.96 19.47 56.29 2.89 1.54 4.58 1.69 1.14 9.74
μ (n ¼ 4) 1.62 3.58 18.97 58.64 3.09 0.69 4.84 1.47 1.05 9.16
σ 0.05 0.27 0.44 1.65 0.15 0.60 0.17 0.17 0.08 0.40
Fabric 2a
8 Gubbacka 0.58 1.20 11.18 75.90 6.79 1.07 3.06 0.73 0.73 5.56
10 Gubbacka 0.63 1.38 12.82 73.27 5.72 0.75 3.49 0.90 0.80 5.97
19 Tallinn 0.78 1.68 14.49 70.58 4.87 0.37 3.77 1.71 0.71 5.90
26 Turku 0.70 1.84 15.80 69.97 4.43 0.27 3.55 0.95 0.75 6.18
16 Tallinn 0.71 1.43 13.84 72.25 5.22 0.48 3.31 1.65 0.89 5.44
24 Turku 0.82 1.70 13.07 72.81 5.57 0.82 3.55 1.31 0.61 5.32
23 Turku 0.78 1.87 15.37 69.53 4.52 0.31 3.84 0.83 0.90 6.57
18 Tallinn 1.01 1.87 13.80 71.03 5.15 1.08 2.86 3.21 0.69 4.45
6 Gubbacka 0.71 2.01 16.12 66.51 4.13 0.58 3.97 1.37 0.98 7.75
μ (n ¼ 9) 0.75 1.66 14.05 71.32 5.15 0.64 3.49 1.41 0.78 5.90
σ 0.12 0.27 1.59 2.66 0.81 0.31 0.36 0.76 0.12 0.92
Fabric 2b
1 Gubbacka 0.71 1.83 16.58 67.23 4.06 0.55 3.89 0.78 0.89 7.53
4 Gubbacka 0.77 1.74 17.38 63.57 3.66 1.33 4.01 0.83 1.05 9.32
2 Gubbacka 0.66 1.74 17.12 62.43 3.65 1.23 3.94 1.11 1.21 10.57
14 Mankby 1.41 2.70 16.37 63.79 3.90 0.98 4.98 1.26 0.83 7.69
17 Tallinn 0.75 1.77 13.99 69.19 4.95 0.44 3.56 2.15 1.19 6.96
9 Gubbacka 0.63 1.32 15.35 63.10 4.11 4.76 3.58 1.49 0.87 8.90
20 Tallinn 0.77 1.85 13.83 71.00 5.13 0.50 3.57 1.52 0.90 6.05
25 Turku 0.85 1.82 14.90 66.96 4.49 0.39 3.42 5.02 0.79 5.85
5 Gubbacka 1.09 1.84 14.22 69.14 4.86 0.43 3.34 1.25 0.93 7.77
μ (n ¼ 9) 0.85 1.85 15.53 66.27 4.31 1.18 3.81 1.71 0.96 7.85
σ 0.25 0.36 1.38 3.14 0.57 1.39 0.50 1.31 0.15 1.53
Fabric 3
21 Tallinn 1.32 2.87 18.00 61.84 3.44 0.40 4.50 1.38 0.98 8.71
E. Holmqvist et al.
Journal of Archaeological Science: Reports 34 (2020) 102627
11
unglazed, six glazed) found from the four Finnish sites, with one additional
unglazed pot found in Tallinn (no. 22). The ceramics in Fabric 1
are made of non-calcareous (CaO < 2.2 wt%), relatively high alumina
and iron clays (Al2O3 = ca. 16–18, Fe2O3 = 5.3–9.5 wt%; see Fig. 7a).
This group includes coarse unglazed sherds (nos. 3, 7, 32), i.e. suspected
“local ware” of Gubbacka and Mårtensby, 4/5 samples chosen from
Mankby, and the Turku samples associated with production waste (nos.
27–30). In addition, both of the clay daub samples from the Gubbacka
furnaces (nos. 35–36) belong to Fabric 1a. The glazed sherds in Fabric 1
can be dated from the 16th to the 18th centuries, whereas the unglazed
sherds show a wide chronological range, possibly starting from the late
Iron Age (no. 32) and continuing until ca. 1800 CE (no. 27).
Fabric groups 2a and 2b include a total of 18 pots, all with glazed
surfaces, found in Tallinn (five), Gubbacka (eight), Turku (four) and
Mankby (one). Compared to Fabric 1, the Fabric 2 ceramics are similarly
non-calcareous (CaO < 4.3 wt%), but display significantly lower Al2O3
values (ca. 12–14 wt%), and higher silica content (above 70 wt% on
average). Although Fabrics 1 and 2 are clearly geologically related, there
are minor differences in certain trace elemental concentrations, e.g.
ZrO2 at ca. 280–430 ppm and Rb2O at ca. 90–130 ppm, compared to
Fabric 1 values of 180–400 and 110–170 ppm, respectively. The Fabric 2
samples are glazed pots dated from the late 14th to the 16th centuries,
thus this fabric group is chronologically and typo-morphologically more
coherent than Fabric 1. This group contains all but one of the pots
analysed from Tallinn, and the floor tile, no. 18.
The cupel (no. 21) found in Tallinn is a clear outlier in the ED-XRF
data set (see Fabric 3 in Figs. 5 and 6), with high Fe2O3, Rb2O, SrO,
ZrO2 and BaO values (Table 2), some of which may be use-derived
contamination or indication of its status as an import.
The fabric assignments of the pots by ED-XRF were confirmed by
micro-chemical SEM-EDS analysis of grain-free ceramic paste/matrix
areas indicative of “pre-temper” clay chemistry, linked to the original
clay source composition (see Holmqvist et al., 2018 and references
therein). The SEM-EDS shows a similar data structure (Fig. 7a–b),
indicating that the fabric groups represent different clay sources. The
SEM-EDS results also confirm that the ceramics were made of noncalcareous
clays (Table 3), with Fabric 1 artefacts (excluding daub
samples nos. 35–36) displaying lower silica–alumina ratios (≤4.0) and a
higher range of MgO content (1.8–4 wt%) than Fabric 2 pots (SiO2 /
Al2O3 = 3.7–6.8; MgO = 1.3–2.7 wt%, Fig. 7b).
In microstructural examination under SEM, the Fabric 1 pots showed
relatively well-sorted grain-size (excluding the heterogeneous daub
material of nos. 35–36) with generously applied sand-temper characterized
by sub-angular-shaped quartz, plagioclase and K-feldspars sized
< 600 µm. Sample nos. 22, 27, 28, 30, 31, 33 and 34 also show large
voids indicative of the use of organic temper. The mineralogy of the
Fabric 1 pots also included frequent biotite mica (< 200 µm), and rare
titanite, ilmenite, apatite, rutile, garnet-group minerals, zircon and iron
oxides (< 100 µm) as natural clay inclusions. The Fabric 2 samples show
similarly applied coarse sand-temper, and largely the same mineralogy
as Fabric 1, but higher occurrence of rutile and iron oxides in the clay
fraction, and only two porous samples (nos. 23–24) with large voids.
Fabric 3 sample no. 21 (ceramic cupel) displays similar general mineralogical
characteristics, but excludes apatite, rutile, or garnet-group
minerals.
The vitrification stage of the clay minerals indicative of the firing
temperatures of the pots was estimated in all of the samples under SEM
using high magnification (1000x) secondary electron (SE) topographic
images (Fig. 8a) to evaluate the level of sintering and glassiness of the
fabrics. The samples showed no signs of vitrification of the paste minerals,
or very early development of vitrification, suggesting that the
firing temperatures remained below or at 800 ◦
C (Bland et al., 2017;
Holmqvist, 2019: 105–107).
The first fabric group is dominated by pots found at the four Finnish
sites, notably representing a wide chronological range (possibly from the
late Iron Age to ca. 1800 CE). Fabric 1 includes stylistically and
morphologically varied artefacts, glazed and unglazed pots, kiln wasters
from Turku, and structural ceramics sampled from Gubbacka as “local”
ceramic material – these factors strongly link the Fabric 1 pots with the
northern side of the Gulf of Finland, i.e. origin in what is now Finland.
This diverse group probably contains objects made in different times and
places. The kiln wasters are of Turku origin, and the same possibly applies
to glazed pots from Mankby and Mårtensby in this group, although
the glazed redwares may have been made in the Helsinki region as well,
Table 4
SEM-EDS measured glaze chemical composition and glaze thickness (t, in µm), order and group assignments by CA dendrogram. All analyses normalized to 100%.
No Date Site t (µm) Na2O (%) MgO (%) Al2O3 (%) SiO2 (%) K2O (%) CaO (%) FeO (%) SnO (%) PbO (%)
Fabric 1a
33 1550–1650 Mårtensby 150–180 0.19 0.10 1.84 31.84 0.28 0.53 0.88 64.33
34 1550–1650 Mårtensby 180 0.27 0.10 2.77 30.11 0.21 0.17 1.25 65.13
15 16th c? Mankby 100 0.41 0.81 4.18 27.59 0.75 0.74 4.09 7.41 54.02
27 1750–1800 Turku 180 0.75 0.74 6.07 35.83 1.16 2.33 3.62 49.49
11 1450–1550 Mankby 180 0.46 0.89 6.00 29.78 0.90 0.70 2.77 58.51
Fabric 1b
31 17th c.? Mårtensby 150 1.15 1.68 11.71 40.75 2.95 0.71 5.20 35.85
Fabric 2a
8 Early 16th c. Gubbacka 80–100 0.26 0.90 4.92 32.82 0.59 0.65 2.89 56.96
10 Early 16th c. Gubbacka 40–60 0.22 0.57 4.61 33.31 0.58 0.69 3.49 56.53
19 Late 14th–early 15th Tallinn 80–100 0.28 1.01 6.87 38.59 0.89 0.92 3.55 47.90
26 1550–1650 Turku 100–180 0.06 0.51 3.64 33.16 0.42 0.45 2.43 59.33
16 Late 14th–early 15th Tallinn 100–200 0.44 0.69 6.63 34.64 0.59 3.66 2.17 51.17
24 Late 15th c. Turku 100–180 0.43 0.86 5.34 33.36 0.75 0.36 1.91 56.98
23 Late 15th c. Turku 100 0.33 0.94 6.26 39.89 1.23 0.39 2.70 48.26
18 Late 14th–early 15th c. Tallinn 80 0.46 6.95 33.63 0.22 2.23 0.74 55.83
6 1450–1600 Gubbacka 100 0.41 0.81 7.89 47.26 1.65 0.98 2.97 38.03
Fabric 2b
1 16th c. Gubbacka 200 0.39 0.26 2.77 27.14 0.31 0.33 4.16 64.90
4 16th c. Gubbacka 200 0.21 0.63 4.73 27.82 0.50 0.35 2.11 63.79
2 c. 1450–1600 Gubbacka 200 0.06 0.51 3.51 27.29 0.31 0.59 0.78 67.17
17 Late 14th–early 15th c. Tallinn 100 0.56 0.75 9.73 56.62 3.06 1.00 2.65 25.63
9 15th c. Gubbacka 60–80 0.17 0.93 4.81 23.99 0.46 0.72 2.27 66.66
20 Late 14th–early 15th c. Tallinn 40 0.06 1.28 7.07 36.70 0.50 1.63 3.25 49.50
25 16th Turku 200 0.46 0.44 4.63 42.06 1.14 1.52 2.04 47.70
5 1450–1600 Gubbacka 300 0.29 0.70 4.92 40.74 0.93 0.97 2.29 49.09
Fabric 3
21 Late 14th–early 15th c. Tallinn 200–250 2.57 1.32 2.10 32.44 2.86 2.82 2.57 53.32
E. Holmqvist et al.
Journal of Archaeological Science: Reports 34 (2020) 102627
12
especially in the 16th–17th centuries (cf. the dates of the glazed Mårtensby
pots in this group). Turku and Helsinki are on the Finnish coast,
only ca. 150 km apart, so pots from these geologically related localities
can be difficult to differentiate based on the geochemical data (see e.g.
Korsman and Koistinen, 1998: 95–97; Al-Ani and Sarap¨a¨a, 2008: 82).
Redware manufacture was positively established in Turku at the turn of
the 16th and 17th centuries, if not earlier (Pihlman, 1989; Wahlberg,
2000; Tulkki, 2003). Nevertheless, the unglazed coarseware pots from
Gubbacka, Mankby and Mårtensby were almost certainly produced in
the greater Helsinki area, e.g. in local village workshops.
One of the Tallinn pots (presumably late 13th c. context) belongs to
this otherwise Finnish-find dominated group, yet its chemical profile in
both the ED-XRF and SEM-EDS data fits very well in this group. Its
tentative status as a Finnish import is supported by its unusual size and
shape compared to typical Tallinn finds. The late 13th century date is
rather early to suggest redware export from Finland; however, we can
speculate that this unglazed pot was produced by Swedish settlers on the
Finnish coast, and traded to Tallinn along with other transactions.
In contrast to the typo-chronologically varied Fabric 1 linked with
the Finnish-coast origin, the other main group, Fabric 2, includes only
glazed sherds dated from the late 14th to the 16th centuries, but the
pottery was found at four sites – Tallinn, Gubbacka, Mankby and Turku –
suggesting that Fabric 2 products were widely distributed. Although this
group includes all the glazed pots and the glazed tile from Tallinn, we
cannot with certainty conclude that these are Tallinn manufacture,
especially considering that there is no direct archaeological evidence of
redware manufacture in Tallinn. Some of the Fabric 2 pots show visual
resemblance to Southern Scandinavian redwares, and this factor,
Fig. 7. a) A scatter plot showing the ED-XRF data silica/alumina ratio vs iron content in the ceramic fabrics, b) SEM-EDS measured silica/alumina ratio vs MgO
content of ceramic pastes; samples marked by CA groupings.
E. Holmqvist et al.
Journal of Archaeological Science: Reports 34 (2020) 102627
13
combined with the close geochemical similarity to Fabric 1 pots (made
on the northern Gulf of Finland), may be evidence that Fabric 2 ceramics
were made in Tallinn by immigrant potters, who also produced pots for
export.
It is apparent from the ED-XRF and SEM-EDS fabric/paste data that
the clays used to manufacture Fabric 1 and 2 pots are geologically
associated, probably from the Gulf of Finland environment. Alternative
interpretations can also be made regarding the source areas of the two
fabric groups. The limited geochemical and mineralogical variation
between Fabrics 1 and 2 could imply that they represent alternative
ceramic recipes originating from the same geological area (modern
Estonia, eastern Sweden, or even the Low Countries?) yet a closer inspection
of the sample distribution in these groups complicates this
scenario. Relative similarity between fabric groups can be expected from
local products of sites located within ca. 180 km from each other (Fig. 1),
especially considering the geological (petrological and geochemical)
similarities between southern Finland (and the Helsinki region, in
particular), and the structural-petrological zone of Tallinn in Northern
Estonia (Soesoo et al., 2004; 2020; K¨ahk¨onen, 2005).
4.2. Glaze composition and application
Of the 34 analysed ceramic sherds, 24 had glaze coating or at least
glaze spots on their surface(s) (Table 1, Fig. 8b). The glaze was usually
applied to the interior and over the edge of the rim, and in some cases to
the exterior of the pot (Figs. 2–3). On visual inspection, the glazes
appear transparent, often yellow hued (Figs. 2–3, 4a, c–d). In the SEMEDS
data on the glazes (Table 4, Fig. 9), no signs of colourants were
detected. It is probable that the yellowish hue derives from ironabsorption
from the Fe-rich ceramic fabrics (see Molera et al., 1997:
29). The majority of the analysed glazes are high lead glazes (defined as
containing 45–60% PbO by Tite et al. 1998), and only three pots (nos. 6,
17, 31) have glazes with PbO content < 45 wt%. The main components
in the glazes are lead oxide (PbO ca. 25–67 wt%), silica (SiO2 ca. 24–56
Fig 8. a) SEM-SE micrograph of ceramic paste of sample no. 25 with early vitrification development; b) SEM-BSE backscatter micrograph of sample no. 26 showing
crystallization on the glaze-ceramic fabric interface.
Fig. 9. A ternary diagram of the SEM-EDS measured glaze compositions. Concentrations are normalized to 100% (the tin oxide concentration of tin-opacified sample
no. 15 is excluded).
E. Holmqvist et al.
Journal of Archaeological Science: Reports 34 (2020) 102627
14
wt%), and Al2O3 (1.8–11.7 wt%), with MgO, K2O, CaO and FeO
(Table 4). Other components do not exceed the 0.5 wt% detection limit
of the SEM-EDS.
Pot no. 15 found in Mankby has a tin–lead glaze applied on its
interior (Table 4), with a typical tin content of ca. 7 wt% (see Tite et al.,
1998: 244; Mason and Tite, 1997), but the glaze still has a high lead
content (PbO = 54 wt%, Fig. 9). The interior glaze of this 16th-century
sherd (Fabric 1a) shows opacity with more brownish/yellowish colour
than the transparent lead glazes (Figs. 2, 4c), but both surfaces of this
sherd are weathered and the glaze is poorly preserved. Tin-opacified
glazes were manufactured e.g. in the Low Countries from the early
16th century onwards (see Mason and Tite, 1997; Carlsson et al., 2018),
but this 16th-century sherd provides no geochemical indication of imported
status, thus it is possible that tin–lead glazes were produced in the
Finnish region at this time.
The glaze thicknesses (Table 4) vary significantly between the
studied samples, and many pots present glazes with very uneven surfaces,
gas bubbles, cracks and abundant (partly or completely undiffused)
mineral grains, undissolved in the firing process (Figs. 10 and 11).
The grains, fairly large in relation to the glaze thickness, often crack the
glaze surface (Fig. 11a, c). These features are characteristic of short
firing times, low temperatures and unstandardized glazing technologies,
starting from the mixture preparation.
High-lead glazes in general have low firing temperatures
(700–1000 ◦
C, see Hurst and Freestone, 1996: 16; Tite et al., 1998), but
the only partly melted glazes analysed here imply inadequate firing
temperatures and times, and firing temperatures below 750 ◦
C leaving
grains undissolved (Molera et al., 2001: 1121). Undiffused, silica-rich
areas in a glaze reduce its transparency affecting its visual appearance,
and can also cause flaking (Tite et al., 1998; al-Saad, 2002: 805).
As a result, the glazes often show weathered exteriors deriving from use
wear or post-depositional damage.
The cupel sample, no. 21 (Fabric 3), shows a glassy matrix filled with
abundant partially melted K-feldspars on its surface (Fig. 11f). The cupels
were made without coating, thus the glassy surface here derives
from the process of bead making at the Sulevim¨agi 4/6 site. There are no
painted patterns visible under the glaze surfaces, and the lead diffusing
into the ceramic fabrics also prevents confirmation of a slip layer applied
below the glaze under SEM (see P´erez-Arantegui and Castillo, 2000;
Krishnan et al., 2005).
The sampled pots systematically present crystals in the glaze-ceramic
fabric interface (Fig. 8b, 10–11) deriving from ceramic fabric reacting
with the glaze mixture during firing, creating chemical diffusion between
the coating and fabric (Molera et al., 2001; De Benedetto et al.,
2004: 618). The crystal phase is characteristic of glazes applied on unfired
ceramic surfaces. This undesired effect reduces the transparency of
the glaze, and can be minimized by optimizing the glaze mixture
(eutectic glaze containing ca. 32% silica, 61% lead oxide and 7%
alumina, see Tite et al., 1998: 251, 253). There is significant variation in
glaze compositions within the ceramic fabric groups (Table 4), suggesting
that different high-lead mixture glazes were applied.
We also examined whether the high-lead glazes were applied as lead
oxide itself, reacting with the clay body during firing (in which case, the
major oxide values of the glaze would match those of the clay body after
subtraction of the lead concentration and re-normalization) – or as a
lead-silica mixture (dissimilar to the clay body chemistry unless the
same clay was used in the glaze mixture; see Tite et al., 1998: 249–249;
Walton and Tite, 2010: 748–749; Palamara et al., 2016: 144–145). The
bi-plots (Fig. 12a–d) of the major oxide concentrations of the glazes
(after subtraction of the lead concentration and re-normalization) versus
those of the clay bodies show that the data of samples 8, 16 and 20 fall
on the unity line, suggesting that for these pots, lead oxide itself was
applied to produce the glaze. This may also apply to sample nos. 19, 24,
and 31, plotting close to the unity line. In these cases, the relict mineral
Fig. 10. SEM-BSE micrographs of Fabric 1 pots with varied glaze thicknesses: a) tin–lead-glazed sample no. 15; b) no. 27; c) no. 33; d) no. 31; showing uneven
glazing (a); bubbles, cracks and undiffused mineral grains, and crystallized fabric-glaze interfaces (a–d); and post-depositional deterioration (c). The well-sorted grain
size and sand-temper are visible in the ceramic fabric of no. 31 (d).
E. Holmqvist et al.
Journal of Archaeological Science: Reports 34 (2020) 102627
15
grains in the glazes may, in fact, derive from the body-glaze reaction.
Apart from no. 31 (being a 17th c. pot of Fabric 1b), pots 8, 16, 19, 20,
and 24 date from the mid-14th to the early 16th century and belong to
Fabric 2 subgroups, associated with Tallinn origin. For the rest of the
pots, the glaze and body compositions appear non-related (Fig. 12), and
a lead-oxide-plus-sand mixture was applied instead.
To conclude, varied glaze compositions and application techniques
are attested in our sample material. It is apparent from the poor physical
quality of the glazes that the pots were not successfully fired, and the
recurrent uneven glaze surfaces and heterogeneous glaze matrices, with
relict grains, bubbles and cracks, are not indicative of consistent application
practices or high-standard glazes (i.e. fully vitrified, glassy
matrices). Perhaps the potters did not aim to produce high-standard
glazes for these modest pots, or did not possess the skills or recipes for
quality glaze making.
5. Conclusions
The results of the ED-XRF and SEM-EDS analyses of 34 redware
objects from five late medieval sites in modern Finland and Estonia, the
hamlets Gubbacka, Mankby, Mårtensby and the towns Turku and Tallinn,
revealed two main compositional ceramic fabric groups, one of
which indicating redware manufacture in the Helsinki region and Turku,
and the other in Tallinn.
Based on the micro-chemical and structural characteristics, the artefacts
were manufactured by applying high-lead-content (PbO < 67 wt
%) glazes on non-calcareous ceramic fabrics prior to firing. The combination
of high-lead glazes with non-calcareous ceramics was invented
already in the 9th–12th century Byzantium with many technological
benefits – high-lead glazes require lower firing temperatures than alkaliglazes,
and are less dependent on the ceramic fabric quality, as their
thermal expansion coefficient is close to that of coarse ceramics (Tite
et al., 1998: 242–253; Walton and Tite, 2010: 754). One of the pots also
had a tin-opacified glaze. In most cases, the glaze was made by mixing
lead oxide with sand (undissolved quartz and feldspar grains), but
sometimes (particularly in the Tallinn group) a lead compound itself was
used. The chosen firing times and temperatures were inadequate to
create glassy coatings, leaving the glazes cracked, filled with grains and
Fig. 11. SEM-BSE micrographs of Fabric 2 glazes, showing different glaze thicknesses, bubbles, cracks and undiffused mineral grains, and crystallized glaze-fabric
interfaces: a) sample no. 6; b) no. 17; c) no. 20; d) 25; e) 26; f) no. 21 (cupel).
E. Holmqvist et al.
Journal of Archaeological Science: Reports 34 (2020) 102627
16
bubbles, and probably jeopardized the impermeability, hardness and the
appearance of the coatings.
Perhaps surprisingly, the analysis did not reveal clear compositional
indications of pottery imported from outside the sample region, that is,
geochemical outliers that could be interpreted as imports from the Low
Countries, eastern Sweden or Germany, all of which are known redware
distributors. However, we are discussing ceramic provenance of a fairly
small sample set, faced with the lack of reference data, and dealing with
artefacts from a period when redware pots were manufactured “almost
everywhere” by mobile craftsmen, small-scale rural and urban producers,
single producers with multiple recipes and forms in their repertoires,
coupled with the potential international impact of Hanseatic
trade (see Carlsson et al., 2018).
In any case, the dominance of redware pots in late medieval contexts
in Finland demonstrates that redwares were very much in vogue not
only in continental Europe and Scandinavia, but also around the Gulf of
Finland, prompting diverse local manufacture of these desired products.
The earliest glazed pots found at the Finnish sites are imports, predating
the “local” glazed pots, hence local potters may have been adapting to
the new glazing technology as a response to these imports. Thereafter,
the lead-glazing practice was continued in the local craftsmanship for
centuries.
Our ceramic samples were prepared as polished cross-sections and
subjected to both ED-XRF and SEM-EDS analyses. While this approach
may have compromised analytical accuracy and sample representation
compared to quantitative analysis using homogenized samples, it was
necessary to control elemental migration from the high-lead glazes into
the ceramic fabrics, which would have largely prevented accurate
geochemical characterisation in bulk analyses. Furthermore, phaseanalyses
under SEM provide technological data on both ceramic and
glaze manufacture that are inseparable characteristics of the late medieval
redware pottery technologies.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgements
The authors would like to thank Georg Haggr´en, Ulla Kadakas,
Marianna Kemell, Jenni Siltainsuu, Kate Sotejeff-Wilson, and Elina
Ter¨av¨a for their collaboration. We also wish to thank two anonymous
reviewers for their valuable comments. This project was funded by the
Svenska kulturfonden, the EU Central Baltic Interreg IV A-Programme,
the Vantaa City Museum, and research grant PRG29 of the Estonian
Research Council, and finished during the first author’s Helsinki Collegium
for Advanced Studies fellowship.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.jasrep.2020.102627.
References
Al-Ani, T., Sarap¨a¨a, O., 2008. Clay and clay Mineralogy. Physical–chemical properties
and industrial uses. Geologian Tutkimuskeskus. M19/3232/2008/41.
Fig. 12. a–d. Bi-plots of SEM-EDS measured Al2O3, SiO2, and FeO concentrations, and SiO2/Al2O3 ratios of the clay bodies and glazes (after subtraction of the lead
concentration and re-normalisation). The data points falling on the unity line indicate that the glaze was applied as lead oxide, whereas in other cases a lead-oxideplus-sand
mixture was used.
E. Holmqvist et al.
Journal of Archaeological Science: Reports 34 (2020) 102627
17
al-Saad, Z., 2002. Chemical composition and manufacturing technology of a collection of
various types of islamic Glazes excavated from Jordan. J. Archaeol. Sci. 29, 803–810.
https://doi.org/10.1006/jasc.2000.0576.
Bencard, M., Roesdahl, E., 1972. Dansk Middelaldertøj 1050–1550. Jysk Arkæologisk
Selskabs Håndbøger I. Nordisk Forlag, Århus.
Bergold, H., B¨ack, M., Johansson, M., Menander, H., Niukkanen, M., Tulkki, C.,
Wallebom, U., 2004. Handled with care: on the typology and symbolism of redware
pottery. Muinaistutkija 2, 2–25.
Bland, C.A., Robert, A.L., Popelka-Filcoff, R.S., Santoro, C.M., 2017. Early vitrification
stages identified in Prehistoric earthenware ceramics from Northern Chile via SEM.
J. Archaeolog. Sci.: Rep. 16, 309–315.
Bliek, G. (Ed.), 1989. Travaux du Groupe de Recherches et d́Etudes sur la C´eramique
dans le Nord-Pas-de-Calais. Actes du colloque de Lille (26–27 mars 1988). S.l. [Lille
& Berck-sur-Mer]: C.R.A.D.C. (Nord-Quest Arch´eologie, num´ero hors-s´erie).
Brorsson, T., 2016. ICP-analys av medeltida och postmedeltida keramik från Lundstr¨oms
plats, J¨onk¨oping, och eftermedeltida keramik från byn Odensj¨o, J¨onk¨oping,
Småland. Arkeologisk rapport 41, Bilaga 5.
Carlsson, K., Forsblom Ljungdahl, V., Gustavsson, J., 2018. Frames of production: The
case of ceramics. Int. J. Historical Archaeol. 22, 365–400. https://doi.org/10.1007/
s10761-017-0422-5.
Caune, A., 2004. Arch¨aologische Zeugnisse von der Bearbeitung des Glases in Riga im 13.
und 14. Jahrhundert, in: Czaja, R., Nawrolska, G., Rębkowski, M., Tandecki, J.
(Eds.), Archaeologia et historia urbana, Pąmięci Tądeusza Nąwrolskiego, Elbląg, pp.
457–464.
Christensen, T., Larsen, A.-C., Larsson, S., Vince, A., 1994. Early glazed ware from
medieval Denmark. Medieval Ceramics 18, Medieval Pottery Research Group,
67–76.
Davey, P., Hodges, R. (Eds.), 1983. Ceramics and trade. The production and distribution
of later medieval pottery in North-West Europe. Papers derived from the Proceedings
of the Medieval Pottery Research Group’s annual conference at Hull, 1980.
University of Sheffield.
De Benedetto, G.E., Acquafredda, P., Masieri, M., Quarta, G., Sabbatini, L., Zambonin, P.
G., Tite, M.S., Walton, M., 2004. Investigation on roman lead glaze from Canosa:
Results of chemical analyses. Archaeometry 46, 615–624.
Elfwendahl, M., 1999. Från sk¨arva till k¨arl. Ett bidrag till vardagslivets historia I Uppsala.
Lund Studies in Medieval Archaeology 22.
Gaimster, D., 1997. German Stoneware 1200–1900. Archaeology and Cultural History.
The Trustees of the British Museum, British Museum Press.
Gaimster, D., 1999. The Baltic ceramic market c. 1200–1600: An archaeology of the
Hanse. Fennoscandia Archaeol. XVI, 59–69.
Gaimster, D., 2007. Redware vessels and tile stoves: signatures of Hanseatic lifestyle in
the Baltic, in: Majantie, K. (Ed.), Pots and Princes: Ceramic Vessels and stove Tiles
from 1400–1700. Archaeologia Medii Aevii Finlandiae XII, Society for Medieval
Archaeology in Finland, Turku, pp. 18–25.
Harjula, J., Helamaa, M., Haarala, J., Immonen, V. (Eds.), 2016. Mankby. A Deserted
Medieval Village on the Coast of Southern Finland. Archaeologia Medii Aevi
Finlandiae XXII. Suomen keskiajan arkeologian seura, Turku.
Harjula, J, Hukantaival S., Immonen V., Ratilainen, T., Salonen, K. (Eds.), 2018.
Koroinen. Suomen ensimm¨ainen kirkollinen keskus. (Turun Historiallinen Arkisto
71) Turun Historiallinen yhdistys, Turku.
Heinonen, T., 2015. Kauppiaita, talonpoikia vai ratsumiehi¨a? 1500–1600-luvun
maaseudun mikroarkeologiaa Vantaan Mårtensbyn tilalla. Vantaan
kaupunginmuseon julkaisuja 32.
Holmqvist, E., 2017. Handheld portable energy dispersive X-ray fluorescence
spectrometry (pXRF). In: Hunt, A.W. (Ed.), Oxford Handbook of Archaeological
Ceramic Analysis. Oxford University Press, Oxford, pp. 363–381.
Holmqvist, E., 2019. Ceramics in Transition: Production and Exchange of Late ByzantineEarly
Islamic Pottery in Southern Transjordan and the Negev. Archaeopress, Oxford.
Holmqvist, E., V¨ais¨anen, R., Koivisto, A., 2014. Redwares from Gubbacka’s Medieval
village: Regional and inter-regional views on ceramic networks and technology
(SEM-EDS). Estonian J. Archaeol. 18 (2), 86–101. https://doi.org/10.3176/
arch.2014.2.02.
Holmqvist, E., Larsson, Å.M., Kriiska, A., Palonen, V., Pesonen, P., Mizohata, K.,
Kouki, P., R¨ais¨anen, J., 2018. Tracing grog and pots to reveal Neolithic Corded Ware
Culture contacts in the Baltic Sea Region (SEM-EDS, PIXE). J. Archaeol. Sci. 91,
77–91.
Hurst, D. and Freestone, I., 1996. Lead glazing technique from a Medieval kiln site at
Hanley Swan, Worchestershire. Medieval Ceramics 20, Medieval Pottery Research
Group, 13–18.
Johansen, P., 1951. Nordische Mission, Revals Gründung und die Schwedensiedlung in
Estland, Kungl. Vitterhets historie och antikvitets akademiens handlingar 74,
Stockholm.
Johansson, M., 2007. Redware vessels in Stockholm. In: Majantie, K. (Ed.), Pots and
Princes: Ceramic Vessels and stove Tiles from 1400–1700. Society for Medieval
Archaeology in Finland, Turku, Archaeologia Medii Aevii Finlandiae XII, pp. 48–58.
Johansson, E., 2013. Younger grey ware: Definition, function and potential. Lund
Archaeol. Rev. 18 (2012), 61–71.
Kadakas., V. and V¨ais¨anen, R., 2012. Padisen sisterssil¨aisluostarin sek¨a Gubbackan ja
Mårtensbyn kyl¨atonttien keramiikkal¨oyd¨oist¨a, in: Russow, E. (Ed.), Padise ja Vantaa:
Keskiajan silta Padisen ja Vantaan v¨alill¨a. Padise Vallavalitsus ja Vantaan kaupunki/
historiatoimikunta, Padise, pp. 323–352.
Kadakas, U., Toos, G., Kalm, A., Nurk, R., Vedru, G., Tasuja, K., Kadakas, V., 2013.
Medieval and early modern suburban site in Tallinn, Tartu Road 1: Contexts and
structures. Archaeological Fieldwork in Estonia 2012. Tallinn, 133–148.
Kilmurry, K., 1980. The Pottery Industry of Stamford, Linconshire, c. AD 850–1250.
British Archaeological Reports. British Series, 84. Archaeopress, Oxford.
Koivisto, A., 2014. Vauras talonpoikaistila. Vantaan Mårtensbyn Lillaksen arkeologiset
tutkimukset, in: Hako, j. (Ed.), Helsingin pit¨aj¨a 2015, Vantaa-Seura, Porvoo, pp.
6–11.
Koivisto, A., Koivisto, R., Hako, J., (Eds). 2010. Gubbacka. Keskiajan arkea Vantaalla,
Museoviraston rakennushistorian osaston julkaisuja 34, Porvoo.
Korsman, K., Koistinen, T., 1998. Suomen kallioper¨an yleispiirteet. In: Lehtinen, M.,
Nurmi, P., R¨am¨o, T. (Eds.), Suomen Kallioper¨a, 3000 vuosimiljoonaa. Suomen
Geologinen Seura Ry, Helsinki, pp. 94–104.
Krishnan, K., Freestone, I.C., Middleton, A.P., 2005. The technology of ‘glazed’ reserved
slip ware – A fine ceramic of the Harappan period. Archaeometry 47 (4), 691–703.
K¨ahk¨onen, Y., 2005. Svecofennian supracrustal rocks. In: Lehtinen, M., Nurmi, P.A.,
R¨am¨o, T. (Eds.), Precambrian geology of Finland – Key to the evolution of the
Fennoscandian Shield. Elsevier, Amsterdam, pp. 343–406.
Lightfoot, E., Naum, M., Kadakas, V., Russow, E., 2016. The influence of social status and
ethnicity on diet in mediaeval Tallinn as seen through stable isotope analysis.
Estonian J. Archaeol. 20 (1), 81–107. https://doi.org/10.3176/arch.2016.1.04.
Mason, R.B., Tite, M.S., 1997. The beginnings of tin-opacification of pottery glazes.
Archaeometry 39, 41–58.
M¨all, J., Russow, E., 2004. Arch¨aologie und Stalinismus – die Anf¨ange der
Stadtarch¨aologie in Estland und die Ausgrabungen auf dem Tallinner Rathausplatz
1953. Zeitschrift für Arch¨aologie des Mittelalters 31 (2003), 145–158.
McCarthy, M.R., Brooks, C.M., 1988. Medieval Pottery in Britain AD 900–1600. Leicester
University Press, Avon.
Molera, J., Vendrell-Saz, M., Garcia-Vall´es, M., Pradell, T., 1997. Technology and color
development of Hispano-Moresque lead glazed pottery. Archaeometry 39, 23–39.
Molera, J., Pradell, T., Salvado, N., Vendrell-Saz, M., 2001. Interactions between clay
bodies and lead glazes. J. Am. Ceram. Soc. 84, 1120–1128.
Niukkanen, M., 2000. Jacoba-kannusta ja jyllantilaispadasta seltteripulloon, in: Uotila,
K. (Ed.) Vesilahden Laukko: linna, kartano, koti. Archaeologia Medii Aevi Finlandiae
IV. Suomen Keskiajan Arkeologinen Seura, Turku, pp. 73–84.
Niukkanen, M., 2007. The form and decoration of redware vessels: functionality and
hidden symbolic meanings, in: Majantie, K. (Ed.), Pots and Princes: Ceramic Vessels
and stove Tiles from 1400–1700. Archaeologia Medii Aevii Finlandiae XII, Society
for Medieval Archaeology in Finland, Turku, pp. 26–35.
Palamara, E., Zacharias, N., Xanthopoulou, M., Kasztovszky, Z., Kov´acs, I., Palles, D.,
Kamitsos, E.I., 2016. Technology issues of Byzantine glazed pottery from Corinth,
Greece. Microchem. J. 129, 137–150.
P´erez-Arantegui, J., Castillo, J.R., 2000. Characterization of red-coloured slips (almagra)
on Islamic ceramics in Muslim Spain. Archaeometry 42 (1), 119–128.
Pihlman, A., 1989. Saviastiat, in: Kostet, J. and Pihlman A. (Eds.), Turun M¨at¨aj¨arvi –
M¨at¨aj¨arvi i Åbo. Raportteja 10. Turun maakuntamuseo, Turku, pp. 83–104.
Pihlman, A., 2010. Turun kaupungin muodostuminen ja kaupunkiasutuksen
laajeneminen 1300-luvulla, in: S¨oderstr¨om M. (Ed.), Varhainen Turku. Raportteja
22. Turun museokeskus, Turku, pp. 9–29.
Pihlman, A. and Savolainen, P., 2019. Ruukuntekij¨an tontti – 1700-luvun savenvalajien
arki historiallisen arkeologian valossa, in: Mustonen, R. and Ratilainen, T. (Eds.),
Pitkin poikin Aurajokea. Arkeologisia tutkimuksia. Raportteja 23. Turun
museokeskus, Turku, pp. 173–184.
Ratilainen, T., 2010. Tiilen k¨ayt¨ost¨a 1300-luvun Turussa, in: S¨oderstr¨om, M. (Eds.),
Varhainen Turku. Raportteja 22. Turun museokeskus, Turku, pp. 31–55.
Ratilainen, T., Harjula J., Immonen, V., Saloranta E., 2016. The medieval town of Turku
– Its roots, founding and development till the 14th century, in: Gl¨aser, M. and
Schneider, M. (Eds.), Lübecker Kolloquium zur Stadtarch¨aologie im Hanseraum X:
Vorbesiedlung, Gründung und Entwicklung, Schmidt-R¨omhild, Lübeck, pp.
559–578.
Russow, E., 2007. Some remarks on medieval and post-medieval redware vessels in
Tallinn, in: Majantie, K. (Ed.), Pots and Princes: Ceramic Vessels and Stove Tiles from
1400–1700. Archaeologia Medii Aevii Finlandiae XII, Society for Medieval
Archaeology in Finland, Turku, pp. 69–80.
Russow, E., 2016. Origines Revaliae – Die ersten hundert Jahre, in: Gl¨aser, M. and
Schneider, M. (Eds.), Lübecker Kolloquium zur Stadtarch¨aologie im Hanseraum X:
Vorbesiedlung, Gründung und Entwicklung, Schmidt-R¨omhild, Lübeck, pp.
537–558.
Russow, E., 2019. Vene asustus Sulevim¨ael, in: Kala, T. (Ed.), Tallinna ajalugu, I. (History
of Tallinn, I: until 1561). Tallinna Linnaarhiiv, Tallinn, p. 211.
Russow, E., L˜ougas, L., Maldre, L., Hiie, S., Kihno, K., Luik, H., Kadakas, V., Sarv, K.,
Kadakas, U., Kalm, A., Reppo, M., 2013. Medieval and early modern suburban site in
Tallinn, Tartu Road 1: artefacts and ecofacts. Archaeological Fieldwork in Estonia
2012. Tallinn, 149–170.
Russow, E., Randoja, K., Bernotas, R., Tvauri, A., Rammo, R., Reppo, M., Ratas, J.,
Kreem, J., L˜ougas, L., 2019. A late medieval treasure trove of Tallinn. Salvage
excavations of the 15th-century landfill between the Jahu and V¨aike-Patarei streets.
Archaeological Fieldwork in Estonia 2018. Tallinn, 185–218.
Salminen, T., 2012. Uusmaalaisten Tallinnanperinn¨ot 1350–1560. in: Russow, E. (Ed.),
Padise ja Vantaa: Keskiajan silta Padisen ja Vantaan v¨alill¨a. Padise Vallavalitsus ja
Vantaan kaupunki/historiatoimikunta, Padise, pp. 183–258.
Saloranta, E., 2010. Puurakentaminen ja puurakennukset Turussa 1300-luvulla, in:
S¨oderstr¨om, M. (Ed.), Varhainen Turku. Raportteja 22. Turun museokeskus, Turku,
pp. 57–77.
Soesoo, A., Puura, V., Kirs, J., Petersell, V., Niin, M., All, T., 2004. Outlines of the
Precambrian basement of Estonia. Proc. Est. Acad. Sci. 53 (3), 149–164.
Soesoo, A., Nirgi, S., Plado, J., 2020. The evolution of the Estonian Precambrian
basement: Geological, geophysical and geochronological constraints. Trans. Karelian
Res. Centre Russ. Acad. Sci. 2, 18–33. https://doi.org/10.17076/geo1185.
E. Holmqvist et al.
Journal of Archaeological Science: Reports 34 (2020) 102627
18
Tite, M.S., Freestone, I., Mason, R., Molera, J., Vendrell-Saz, M., 1998. Lead glazes in
Antiquity – Methods of production and reasons for use. Archaeometry 40 (2),
241–260.
Tulkki, C., 2003. Oliko keramiikka-astioiden valmistus mahdollista uuden ajan alun
Turussa?, in: Sepp¨anen, L. (Ed.), Kaupunkia pintaa syvemm¨alt¨a: arkeologisia
n¨ak¨okulmia Turun historiaan. Archaeologia Medii Aevi Finlandiae IX, Turku, pp.
207–222.
V¨ais¨anen, R., 2016. Kolme kyl¨a¨a. Vantaan keskiaikaisten ja uuden ajan alun kyl¨atonttien
arkeologiset tutkimukset 2008–2014. Vantaan kaupunginmuseon julkaisuja 33.
Wahlberg, J., 2000. Keraamisen aineiston PIXE-analyysi. Laukon kartanon, Kuusiston
linnan ja Aboa Vetus-museon n¨aytteet, in: Uotila, K. (Ed.), Vesilahden Laukko: linna,
kartano, koti. Archaeologia Medii Aevi Finlandiae IV, Suomen Keskiajan
Arkeologinen Seura, Turku, pp. 123–129.
Wahl¨o¨o, C. 1976. Keramik 1000–1600 i Svenska fynd. Archaelogica Lundensia.
Investigationes de antiqvitatibus urbis Lundae, VI. Kulturhistoriska museet, Lund.
Walton, M., Tite, M.S., 2010. Production technology of Roman lead-glazed pottery and
its continuance into Late Antiquity. Archaeometry 52 (5), 733–759. https://doi.org/
10.1111/j.1475-4754.2009.00506.x.
E. Holmqvist et al.