The story of two historical textile fragments: Technical analysis and reconstruction of the lost textile pattern

Abstract

Cultural heritage textile artifacts that have been preserved only in their smaller fragmented remains represent a very interesting and complex matter, very rarely analyzed and studied individually as a separate entity. This case study covers multidisciplinary research conducted on two 18th century silk fragments with a seemingly identical pattern. Demonstrating the possibilities and challenges which arise when analyzing and interpreting incomplete cultural heritage textile materials, this study can serve as a stimulus for creating a large-scale database of historical fabrics which would allow comparison based on their differences or common characteristics. By applying only non-destructive and micro-analytical methods it has been determined that the fragments were made of the same material and using the same manufacturing techniques. It has been proven that fragments were not part of the same historical fabric, but most likely originated from different parts of the same liturgical vestment. The reverse engineering process applied to preserved fragments has resulted in a detailed technical documentation and a complete reconstruction of the lost original pattern.

Keywords

Historical textiles textile fragments pattern reconstruction technical analysis multidisciplinary research

Textile production is one of the most ancient crafts, dating back to the very beginning of mankind. Owing to the presence of textiles and clothing in many aspects of human life, every historical fabric is a valuable and fragile document of the development of human activity that influenced its origin and its use.¹ What is preserved today, however, are often historical textiles in a fragmentary state that nevertheless represent a very valuable source of information and can allow a potential reconstruction of what the original garment once looked like.²

In the case of historical textile fragments where very little of the original textile material is preserved, the challenge is to determine how to read all the information needed for a detailed technical and stylistic analysis. Scientific research usually focuses on valuable historical textile objects that most often belong to some historical figure, whereas such examples of a small and unknown fragmented textile are rarely researched and processed in detail.

Two intriguing silk fragments with a seemingly identical colorful floral pattern were found among valuable textile artifacts stored within the textile conservation workshop at the Department for Art and Restoration, University of Dubrovnik.

Owing to the lack of written sources about the date, origin, technology and technique of manufacture, as well as information about the original appearance and purpose of these two 18th century textile fragments, a number of questions remain open.

The knowledge of when and where a textile was produced, together with other historical data available, is essential to its understanding.

The city of Dubrovnik, located in the southern part of the Republic of Croatia (Ragusa), is a large cultural and historical monument, which held the title of a United Nations Educational, Scientific and Cultural Organization (UNESCO) World Heritage Site since 1979. Between the 14th and 19th centuries the Republic of Dubrovnik ruled itself as a free state. A very favorable geographical position enabled it to be the link between the East and the West. Until the second part of the 19th century Dubrovnik played a significant role as a very important center on the route of a wide merchant network of textile and natural dyes. Although it did not have its own industrial textile production, it is widely considered by historians to be one of the most important shipping, trading and cultural centers.³

In the 18th century colorful silk patterns of interlaced exotic flowers and fruits, still life, architectural elements and landscapes prevailed in textile art.⁴ In his book Le dessinateur pour les fabrique d’etoffes d’or, d’argent et de soie from 1765, Nicolas Joubert de l’Hiberderie considers the weaving of three-dimensional exuberant flowers and fruits to be equal to the quality of paintings.⁵ Famous French silk designers such as Jean Revel (1684–1751) and Philippe de Lasalle (1723–1804) played key roles in textile production by developing creative technical solutions and sophisticated techniques for weaving complex and naturalistic silk designs of that era. For this reason we are not surprised by the fact that the most valuable preserved textile examples of this period are generally attributed to the French textile manufacturers, in particular to the textile production in the city of Lyon.^4–7

By the second half of the 18th century figured silks intended for liturgical usage were woven with Christian symbols incorporated in the design. Ecclesiastical motifs like wheat ears and vine—iconography of bread and wine—became very popular well into the 19th century. Although these Christian symbols occur in many historical vestment silks, it is often quite difficult to detect them among the prevailing flower decorations and therefore recognize them as a silk woven for church use.⁸

Expensive silk materials have often been reused and readapted throughout history to create new textile objects or repair already damaged ones.⁴ A wide range of uses for patterned silk fabrics can be reflected in the preserved historical objects: they were used for the production of costumes, liturgical vestments and interior decoration (e.g. curtains, wall hangings, upholstery fabrics), as well as many other objects.^4–10

The aim of this paper is to share our research findings, but also to demonstrate that researching fragmented textile materials represents a great challenge because of limited information available, as well as the delicate nature of the material itself. It also aims to prove that the analysis of historical, stylistic and technological aspects, as well as their correct interpretation, requires multidisciplinary interaction of different scientific fields that enables better understanding of obtained data and the reconstruction of lost information.¹¹ Finally, the goal of the paper is to show that by revealing all the hidden facts one can fully understand and appreciate the beauty, complexity and importance of preserving textile heritage.

Experimental research

General description

Two silk fragments with supplementary pattern wefts in various colors (Fragment A, 20 cm × 16.7 cm (Figure 1) and Fragment B, 19.7 cm × 13 cm (Figure 2)) were acquired cut in an irregular rectangular shape, with selvages preserved on both materials. Both textile fragments originally belonged to the local art restorer Ivo Scattolini (1872–1945), who collected many samples of historical fabrics during the first half of the 20th century, mostly from liturgical vestments that were no longer in use. After his death, the entire textile collection was inherited by his daughter, who then passed the fabrics into the hands of Danijela Jemo, the head of the textile conservation workshop, who added them to the valuable textile collection treasured at the University of Dubrovnik.

Figure 1.

Fragment A: (a) front; (b) reverse.

Figure 2.

Fragment B: (a) front; (b) reverse.

Condition assessment

Both fragments are in relatively good condition, which is why it was possible to analyze and investigate them more thoroughly.

The previous owner kept textiles in her possession stored in the attic, exposed to the damaging microclimatic conditions. Moreover, she tended to cut fabric samples from the remaining textiles, in order to frame them or give them away as gifts. According to her testimony, her father received these valuable but much damaged textile objects and fabrics from the church, after they became unusable.

The storage conditions in our churches are often inadequate. Items are stored in wooden cabinets and chests, in locations subjected to the effects of many degradation mechanisms and microclimate oscillations. We can only assume what caused them to become unusable, but it is a given fact that historical textile materials deteriorate over time. Inappropriate handling or repairs can further accelerate their degradation.

A significant chromatic difference is visible between the two textile fragments: Fragment B has a slightly darker (brownish) tone, whereas Fragment A is of a brighter color. This occurred following cumulative exposure to electromagnetic radiation of visible light that causes physical and photochemical degradation of textile fiber properties, especially to silk.¹² The degree of discoloration is particularly evident when comparing the folded edge of Fragment B that preserved the original color from the deleterious effect of long-term light radiation.

As Fragment A does not contain folded parts, it is not possible to determine the originality of the colors or compare them with those preserved on Fragment B. The exposure of silk fragments to inadequate microclimatic conditions and storage methods has undoubtedly accelerated the process of material degradation. Both fragments show signs of past usage and mechanical damage—surface wear, torn structure and partial or complete loss of multicolored pattern wefts (Figure 3).

Figure 3.

Loss of floating multicolored pattern wefts: (a) complete; (b) and (c) partial.

Research methodology

The choice of the analytical method depends on the material content and its current state of preservation. The methodological approach to the analysis of such textile material should be a synthesis of interdisciplinary and comparative research, which implies the study of historical sources, conservation–restoration research and technological analysis (Figure 4).

Figure 4.

Planning process of textile cultural heritage research applying a logical sequence of tasks to generate and collect knowledge and data about the artifact being investigated.

Technical documentation of important structural parameters is of great interest because it provides important insight into technical specifications of historical fabrics, enhances dating and attribution methods and provides general knowledge on historical weaving techniques and technology.¹,¹³,¹⁴

Textile fiber microscopy was applied for the identification of fibers used in the weaving of fragments, according to their specific visual characteristics.¹⁵ The aforementioned analysis required sensitive micro-sampling techniques in order to preserve the structural and aesthetic integrity of the material, complying with the code of ethics and the guidelines of the conservation–restoration profession. A trinocular light microscope Olympus BX40F4 microscope with translucent and reflective light, connected to a computer with Olympus SC30 camera and Stream Start (v1.5.1-8521) software, was used for data analysis and digital storage.

Structural parameters of silk fragments such as weave structure and yarn thickness were compared by scientific analytical methods in order to verify the similarities and diversities of the examined materials. The analysis was performed using Dino Lite Pro AM413T5 (1.3M resolution, 500× magnification) and Dino Capture 2.0 software (v.1.5.18A).

Graphic design software Adobe Photoshop CS6 (v.13.0.1) was used for pattern reconstruction and ArahWeave software for subsequently recreating the weave draft. A different range of Photoshop tools provided accurate visual correction of distortions and adjustment of dimension and color values and also enabled us to merge scanned materials into one complete image.

An unprofessional approach and/or improper actions when analyzing fragile and valuable historical textiles may cause further damage. According to ethical guidelines and the principles of the conservation and restoration profession, it is of vital importance that the analytical methods and object sampling are carried out by an art conservator or an equally trained professional.¹,¹³,¹⁴

Pattern comparison

The aesthetic features suggest that both fragments date back to the 18th century. Visual comparison and interpretation of individual pattern elements and its composition revealed that both fragments contain almost identical floral motifs. The shared pattern contains characteristic vines, flowers and leaves placed in a symmetrical composition with a colorful flower bouquet motif placed in the central field, a very common and widespread design template of 18th century patterned fabrics.

Typical Christian symbols of the Eucharist, such as wheat ears and grapevines that traditionally represent the body and blood of Christ, have also been detected on silk fragments (although Fragment B does not cover the wheat ear motif because the yellow patterned area is not preserved). Considering the proven connection of the Scattolini family with church art, this would support the theory that the silk fragments were indeed originally parts of a liturgical object, but this hypothesis cannot be fully proved in the absence of written or photographic documentation.

Detailed inspection of textile fragments revealed that there is indeed a matching correlation in preserved pattern elements, but also some differences and deficiencies of its execution. The corresponding floral motifs on silk Fragment B are slightly larger and sharper in comparison with those of Fragment A. Absence of certain leaf elements that was not caused by physical degradation or loss of floating pattern weft has also been found (Figure 5).

Figure 5.

Exclusion of leaf motives within pattern repeat: (a) Fragment A; (b) Fragment B.

A very important aspect of textile design is the pattern composition and orientation. Most 18th century samples show two types of pattern composition: two pattern units placed in a straight repeat or in a point repeat (Figure 6).⁴,⁵ For this purpose, preserved selvages were an important factor as they indicate the borders of the original pattern unit. The preserved parts of silk Fragment A revealed that this pattern design falls into the latter category, but in a half-brick repeat composition.¹⁶

Figure 6.

Typical pattern design composition of 18th century historical fabrics: (a) straight repeat; (b) point repeat.

The width of the pattern sample unit preserved on silk Fragment A has been measured (13.3 cm) and it allowed us to reconstruct the full-size of the pattern repeat (26.6 cm), which corresponds to the dimensions described by Joubert de l’Hiberderie in the 18th century.⁴,⁵ In her attempts to systematize different weave widths of 18th century historical fabrics, without selvages, A. Jolly, a textile historian, states that written sources document standard sizes ranging from 49.5 to 58.3 cm.⁴ This would imply that our silk fragments were once part of fabrics that accommodated two complete pattern repeat units across their width. Historic fabrics made with a fabric width of 53.2 cm, as in our case, would therefore correspond to the strict regulations and parameters specified for Venetian textile products of that time.

Furthermore, it has been discovered that the preserved parts containing brown and light blue pattern weft yarns (i.e. the upper and lower edge of the silk fragments) align perfectly together and form identical floral patterns, thus opening the possibility for complete reconstruction of the original pattern design. Both fragments proved to be crucial for successful reconstruction of the repeating pattern cycle as one fragment contains a completely preserved area that is missing on the other sample, and vice versa.

Thread comparison

Sometime during the 1730s, Jean Revel, a famous French silk designer, developed a specific technique used for weaving hyper-realistic patterns, called point rentrés or berclé. It uses two opposing wefts of different colors and interweaves them gradually and randomly to produce smooth color transitions and shading effects for a particular motif.⁹ This recognizable patterning method has not been detected on Fragment A and Fragment B, which might indicate that these textile materials were produced in the latter half of the 18th century.

The impression of spatial depth and vibrancy of woven motifs was instead created in this case by the technique of using pattern wefts made of two or multiple different colored fibers that visually create the three-dimensional effect of the main pattern. (Figure 7).⁹,¹⁰ The colorful pattern is consistently made of green pattern wefts in combination with brown, light blue, dark blue, pink, orange red, violet and/or yellow pattern wefts, in proportion of 1:1. It is also worth noting that microscopic and microchemical tests have shown that both fragments are entirely made of multifilament silk fibers.

Figure 7.

Multicolored pattern wefts: (a) in combination with ground weave; (b) blue and green multifilament yarns.

When comparing the results acquired by microscopic and structural analysis of Fragment A and Fragment B, the obtained values almost completely coincide, down to the combination of different colored fibers in the composition of used pattern wefts. Image analysis for non-destructive measuring of the yarn thickness was performed using Dino Lite Pro AM413T5 digital microscope.¹⁷ Yarn thickness was measured in five different places of the same yarn sample. The calculated basic statistical values for measuring yarn thickness (arithmetic mean ( $\bar{x}$ ), standard deviation and coefficient of variation) are represented in Table 1. The maximum deviation (coefficient of variation) for Fragment A is for pattern weft thickness and that is 3.95%; the maximum deviation for Fragment B is also for pattern weft thickness, 3.60%. The minimum deviation was for ground warp thickness for both fragments.

Table 1.

Results of measuring yarn thickness with Dino Lite microscope under 195× magnification

	Ground warp thickness (mm)		Ground weft thickness (mm)		Pattern weft thickness (mm)
Measurement number	Fragment A	Fragment B	Fragment A	Fragment B	Fragment A	Fragment B
1	0.170	0.171	0.191	0.214	0.258	0.241
2	0.175	0.172	0.195	0.212	0.238	0.234
3	0.178	0.179	0.201	0.199	0.237	0.247
4	0.179	0.172	0.203	0.206	0.254	0.243
5	0.179	0.169	0.201	0.212	0.241	0.258
Mean (mm)	0.1762	0.1726	0.1982	0.2086	0.2456	0.2446
SD (mm)	0.0038	0.0038	0.0050	0.0061	0.0097	0.0088
CV (%)	2.16	2.20	2.52	2.92	3.95	3.60

Weave structure comparison

The two fragments are both woven in damask with supplementary pattern wefts, alternating in various colors (damas lancé a trame alternatif).¹⁸ Damask is a type of self-patterned fabric in which the ground weave is created by the warp-faced and the pattern by the weft-faced version of the same weave structure or by using two different but compatible weaves.¹⁹ The earliest forms of damask technique appeared in ancient China during the Han dynasty (from 206 BC to AD 220), yet its origin has been largely attributed—as the name suggests—to the ancient city of Damascus, in today’s Syria.²⁰,²¹

The optical illusion of the multilayered floral pattern was achieved by using damask weave as a lustrous ground structure for colored supplementary wefts. The white damask pattern follows the shape of the main multicolored pattern, yet also includes some additional pattern elements. It has been documented that historical patterned fabrics from the 18th century onwards differ not only by stylistic features, but also by weaving techniques. Damask weaves with supplementary weft patterning prevailed until the period of Naturalism in the 19th century when other weave structures, such as lampas, become more dominant.⁴

Both silk fragments are woven in classical satin damask, 5-end weft-faced satin pattern on a 5-end warp-faced ground, with pattern wefts woven with the ground warp in combination of 10-end and 20-end satin weaves (Figure 8). The weaving structures differ only by being woven in a mirrored state.

Figure 8.

Detail of Fragment B reconstructed weave structure.

In some cases, every binding point of the weave structure would be mirrored along the central axis of the pattern in order to achieve perfectly symmetrical patterns. This weaving technique, however, was not applied on silk Fragment A, which has the central part of the pattern preserved. It leads us to the conclusion that the weaving structure was consistent across the entire original fabric and indicates with great certainty that these silk fragments were not part of the same fabric.

Despite that fact, the technical properties of the silk fragments, such as density, material and color distribution, are nearly identical (Table 2).

Table 2.

Overview of technical properties between Fragment A and Fragment B

		Fragment A	Fragment B
WARP	Main warp
	Density	92 yarns/cm	94 yarns/cm
	Material	Silk	Silk
	Color	White (A)	White (A)
WEFTS	Main weft
	Density	40 yarns/cm	40 yarns/cm
	Material	Silk	Silk
	Color	White (a)	White (a)
	Pattern wefts
	Density	40 yarns/cm	40 yarns/cm
	Material	Silk	Silk
	Color	Green (b)Dark blue (c) Light blue (d) Brown (e) Yellow (f) Violet (g) Orange red (h) Pink (i)	Green (b) Violet (c) Orange red (d) Pink (e) Dark blue (f) Light blue (g) Brown (h)
	Weft sequence	7^a(1a 1b 1a 1c) 50(1a 1b 1a 1d) 81(1a 1b 1a 1e) 108(1a 1b 1a 1f) 64(1a 1b 1a 1g) 82(1a 1b 1a 1h) 25^a(1a 1b 1a 1i)	55^a(1a 1b 1a 1c) 67(1a 1b 1a 1d) 133(1a 1b 1a 1e) 52(1a 1b 1a 1f) 50(1a 1b 1a 1g) 47^a(1a 1b 1a 1h)

^aIncomplete count due to fragment cut.

The technical information provided in Table 2 can reveal other hidden elements of the silk fragments. For example, by multiplying the obtained warp density values and the aforementioned presumed fabric width (53.2 cm) we can estimate the total number of warp ends used, which would roughly equal between 4894 and 5000 warps needed to weave the original fabrics, excluding the selvages.⁴,¹⁶

Both silk fragments have two corresponding colored areas (i.e. patterned sections within the weft sequence) completely preserved, one created by light blue pattern wefts and the other created by orange red pattern wefts. Interestingly, the sizes of light blue patterned areas on both fragments match perfectly, where both samples contain a total of 50 light blue pattern wefts, yet the central motif of an orange red flower shows a significant difference in its height. Close inspection revealed that the orange red patterned area measures 3.7 centimeters on Fragment A and 3.3 centimeters on Fragment B. The dimensional deviation has been confirmed also by counting the number of orange red pattern wefts used for creating this pattern, where it was found that Fragment A contains 82 and Fragment B contains 67 orange red pattern wefts within that corresponding area (Figure 9). Due to the difference in height, certain motifs therefore appear understandably somewhat elongated on Fragment A. This had to be taken into account and compensated for later when applying digital image methods to pattern reconstruction.

Figure 9.

Weft sequence and number of pattern weft picks within the colored areas.

A small pattern error on Fragment A has been discovered, where a brown instead of a yellow supplementary pattern weft was woven in, which suggests a miscalculation in weft counting, probably occurring during the hand-weaving process (Figure 10).

Figure 10.

Pattern error under the pattern motif on Fragment A.

The preserved selvages of Fragment A and Fragment B were also analyzed and compared. They continue in the same weave structure as the rest of the fragments, but differ in width and color. The selvage on Fragment A is wider, made with white and blue warp ends, whereas the selvage on Fragment B uses a combination of red and white warp ends (Figure 11). By counting the warp ends it has been found that a different amount and proportion of white and colored warps were used in production, which made the width of Fragment A selvage larger by 0.9 mm. The appearance and properties of the compared selvages further suggest that these two textile materials were derived from different fabrics.

Figure 11.

Images of preserved selvages: (a) Fragment A; (b) Fragment B.

Digital pattern reconstruction

The application of graphic software for photographic processing is a very important tool in the documentation and reconstruction of lost pattern and weaving drafts.²² Certain computer programs allow virtual simulation of complex weaving structures that can provide an interesting educational contribution to museum collections or publications and contribute to the general understanding of historical techniques, weave structures and appearance of historical textiles.²³ Virtual conservation enables precise processing of aesthetic imperfections and faults on textile objects, without the necessity of direct interventions on the material itself.²

Program Adobe Photoshop CS6 (v.13.0.1) has been primarily used to create control grids and guides, digitally correct visual distortions, adjust dimension and color values and merge scanned materials into one complete image. The damaged areas have been virtually restored by replacing them with those in preserved parts. It should be emphasized again that due to dimensional deviations of certain pattern areas the reconstructed samples have both been resized to their joint central value.

All pattern elements of the silk fragments (damask pattern, multicolored floral pattern) have been isolated and divided into separate digital layers in order to achieve better visibility and the ability to control individual pattern elements independently. The final appearance of the reconstructed pattern has been obtained by multiplying the repeating unit according to the aforementioned defined rules of pattern composition (Figure 12).

Figure 12.

Virtual simulation of the reconstructed pattern repeat.

The detailed technical analysis and reconstructed weave draft make it possible for reproducing historical textiles on a modern jacquard loom. The reconstructed pattern repeat unit was used as a template for creating weave draft with specialized software. For this purpose, ArahWeave, a computer-aided design/computer-aided manufacturing textile software for jacquard woven fabric design, was used together with ArahPaint for drawing and image editing in seamless repeat, and ArahView 3D was used to present fabric on the three-dimensional model (Figure 13).

Figure 13.

Reconstructed weave draft of Fragment A using ArahWeave computer-aided design/computer-aided manufacturing software: (a) complete; (b) detail.

Discussion

The choice of analytical methods and the way in which they are implemented is largely influenced by the current state of the historical textile. When the material is in a degraded condition, handling and research are difficult and require a special approach. In this case, due to the size of the fragments, a very small amount of material was available. Nevertheless, significant information and data have been extracted and some unexpected details important for understanding the construction and history of the object have been revealed.

Different surface coloration indicated inadequate past preservation methods (see the section on Condition assessment), yet it also hindered us from visualizing the original appearance and from determining the similarities between the silk fragments without additional physical and chemical analysis of the dye composition.

Great similarities of technical parameters, such as thickness, color and material of the yarns used, as well as the final appearance of the reconstructed pattern, suggest that the silk fragments are indeed parts of the same historical fabric; however, some glaring mismatches raise certain questions for debate.

Both textile fragments are made in the same damask weave with supplementary multicolored wefts, but in a mirrored version. The warp density in both silk fragments is nearly identical. The difference in weaves and in the subtlety of pattern execution could not be justified by a possible point tie-up because the weave structure does not follow the point repeat of the pattern. The inconsistent pattern height and lack of individual motifs may have occurred during an inconsistent hand-weaving process. On the other hand, no deviation has been detected in the width of the pattern so it is possible to conclude with great certainty that the fabric width of both silk fragments was originally identical. Consequently, it is possible to conclude that these historic fabrics contained two complete repeat units across their fabric width, as shown on the reconstructed pattern repeat.

The irrefutable argument that distinguishes the two silk fragments, however, is the appearance and properties of their preserved selvage. Despite the well-preserved written sources and regulations about textile production throughout history, in which different colored selvages can be made as an indicator for correct orientation and use, numerous irregularities have been discovered among preserved historical fabrics. Although numerous historical and technological research studies show great interest in selvage properties, cloth width and total number of warp ends of historical fabrics, these parameters have not yet shown convincing results for determining the origin of production.⁴,²⁴ Therefore, it is also desirable to interpret cautiously the relevance of associating the reconstructed width of the silk fragments from Dubrovnik with the documented Venetian standards.

Nonetheless, it is evident that the pattern composition and motifs of the two preserved textile fragments match almost perfectly. The possibility of a very faithful reproduction exists, as it was common practice to copy earlier works in textile industry during later centuries, but that hypothesis seems rather unconvincing. Considering the suggested time period of production, the presumed weaving technology used on two silk fragments was one of the forerunners or an early type of jacquard loom. During the 18th century the drawloom was undoubtedly brought to its highest sophistication and the effort was now concentrated in finding easier and less time-consuming methods for weaving intricate patterns and possibly eliminating the need for a drawboy. The development of the jacquard loom involved essentially the perfection of a precise and reliable mechanism to control the opening of a useable shed, using punched cards for the patterning. The punched card method, in combination with a system of needles and hooks, was invented as early as 1725 by Basile Bouchon but it did not obtain useful results. In 1728 Jean-Baptiste Falcon succeeded in improving the system by using a pasteboard instead of Bouchon’s paper and constructed an apparatus consisting of simple cords, leashes and metal hooks. Each card represented one row of the pattern and the entire set of cards for weaving the pattern unit was sewn together and hung over a wooden prism placed above. However, Falcon’s loom was never generally used as it presumably did not function with sufficient precision and a drawboy still had to be employed. The next technological step came in 1745 from another French inventor, Jacques de Vaucanson, who constructed the first completely automated loom. His version could be operated by the weaver himself without the assistance of a drawboy. Vaucanson also utilized punched cards and took them over a barrel placed uppermost on the loom which could be moved stepwise by the weaver with a long treadle. This loom likewise never obtained any practical success.²⁰ However, all these previous attempts allowed Joseph-Marie Jacquard to study and succeed in 1801 to construct the earliest programmable mechanical loom for weaving complex patterned fabrics that in its core has not significantly changed to this day.

The original use of this type of fabric is still unclear, but, based on the composition of the reconstructed pattern, the used motives of wheat ears and grapevines and the knowledge of the activities of the previous owners, one can only assume that the fragments were once part of the same liturgical vestment. This is supported by the fact that liturgical vestments may come as a set of different objects, such as cope, chasuble, burse, maniple and chalice veil, in different dimensions and shapes but made of the same material and using the same manufacturing techniques.

Comparison with similar historical samples proved to be a useful method for dating and attribution of the silk fragments; however, it should be emphasized that there is insufficient scientific literature and databases dealing with historical textiles, which represents a major obstacle for advanced research and comparative analysis of fragile textile cultural heritage. Creating a digital textile archive of fabric designs can be an important and valuable source for textile researchers, related industries and students.²⁵,²⁶

Conclusion

An interdisciplinary approach to research and interpretation of enigmatic silk fragments from Dubrovnik involved different historical and technical aspects that all provided information which otherwise would not have been revealed by separate analytical methods.

Microscopic analysis of the yarns has been used to identify the type and surface morphology of the used fibers. The examined samples have shown that both fragments were made entirely of silk. Technical analysis and comparison of the fragments revealed that their thread and weave structure are also virtually identical. Historical and stylistic analysis determined the characteristic features and associated them with those of 18th century historical fabrics. Visual examination of the silk fragments has revealed existence and correlation of identical design elements. Although the clarity and continuity of the pattern sample was limited by the dimensions and damages of the silk fragments, it was still possible to fully reconstruct the original appearance of their technical repeating unit by understanding the weaving technology and pattern design of historical fabrics. All given data suggest that silk fragments were made in the same historical period and that they were very likely produced by the same textile manufacturer or workshop.

The reconstruction of the pattern repeat unit required digital image processing to mitigate existing pattern disproportions and to piece together all the different parts from both samples. As a reverse engineering process of preserved historical fabrics, a reconstructed pattern repeat unit with documented technical parameters can potentially serve as a template for research or reproduction, should such need or interest arise in the future.

The knowledge and experience gained in this research represent a valuable basis for comparing other similar fragments within our textile collection. Future research projects will be focused on developing multidisciplinary research tools and user-friendly practical methods for analysis and technical documentation of historical patterned textiles. This will be a precondition for establishing a cloud-based digital database of patterned historical fabrics, consisting of a historical analysis and a reconstructed weaving draft with detailed technical specifications. The obtained information could prove useful for researchers of various fields of textile science. An established large-scale database would allow widely available comparative stylistic and technical analysis of historical fabrics based on their differences or common characteristics, enabling worldwide educational and scientific data exchange. It would greatly enhance the existing methods of determining a fabric’s authenticity and attribution and show the correlation and geographic distribution of certain historical textile patterned fabrics, as well as potentially allowing digital reconstructions and reproductions using modern weaving or printing techniques.

In the textile conservation field such techniques can enable effective solutions of reintegrating the missing parts of original historical fabric with recreated materials. In some cases, complete facsimiles of fragile and degraded original textile objects can serve as replacements in museums and other public places where they are exhibited. Meanwhile, the original item can be stored in a safe condition and its deterioration can be reduced to a minimum extent, thus preserving the aesthetical values and visual appearance.

As shown in this case study, even small and damaged historical textile fragments may provide valuable information that can complete, unlock and revive incomplete or seemingly lost mosaics of rich and diverse historic fabrics, adding value to the preservation of textile cultural heritage.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Željko Penava

Danijela Jemo

References

Phipps

Looking at textiles: a guide to technical terms. Los Angeles: Getty Publications, 2011, pp.6–12.

Cybulska

To see the unseen: computer graphics in visualisation and reconstruction of archaeological and historical textiles. In: Mukai

(ed.) Computer graphics. InTech Rijeka, 2012, pp.213–228.

Jemo

Parac-Osterman

Đ.

Production, technology, dyeing and textile trading in the context of maritime-trade routes of Dubrovnik. Naše more 2016; 63: 36–43.

Jolly

Seidengewebe des 18. Jahrhunderts II. Naturalismus. Riggisberg: Abegg-Stiftung, 2002, pp.16–43.

Miller

LE.

Representing silk design – Nicolas Joubert de l’Hiberderie and le dessinateur pour les étoffes d’or, d’argent et de soie (Paris, 1765). J Des Hist 2004; 17: 29–53.

Hafter

DM.

Philippe de Lasalle: from mise-en-carte to industrial design. Winterthur Port 1977; 12: 139–164.

Markowsky

Europäische Seidengewebe des 13.-18. Jahrhunderts. Kunstgewerbemuseum Köln, Köln, 1976, pp.27–30, 64–66, 93–98.

Johnstone

High fashion in the church. Leeds: Maney Publishing, 2002, p.90.

Jolly

Fürstliche Interieurs: Dekorationstextilien des 18. Jahrhunderts. Riggisberg: Abegg-Stiftung, 2005, pp.9–53.

10.

Tietzel

Geschichte der Webkunst: technische Grundlagen und künstlerische Traditionen. Köln: DuMont Buchverlag, 1988, pp.181–184.

11.

De la Torre

Assessing the values of cultural heritage: research report. Los Angeles: Getty Conservation Institute, 2002.

12.

Nilsson

Vilaplana

Karlsson

, et al. The validation of artificial ageing methods for silk textiles using markers for chemical and physical properties of seventeenth-century silk. Stud Conserv 2010; 55: 55–65.]

13.

Kodrič Kesovia

Jemo

. Konstrukcijska analiza veza u tkanju tekstilnih materijala u dijagnostičkom istraživanju umjetničkih predmeta.

The Preservation of Cultural Heritage in Croatia Annual, Zagreb: Ministry of Culture of the Republic of Croatia, 2014; 36: 163–172.

14.

Finch

Textiles as documents of history and those who care for them. In: Brooks

(ed.) Textiles revealed: object lessons in historic textile and costume research. London: Archetype Publications Ltd, 2000, pp.7–9.

15.

Greaves

Saville

BP.

Microscopy of textile fibres. BIOS Scientific Publishers Limited, Oxford, 1995, pp.5–15.

16.

Sonday

Pattern and weaves: Safavid lampas and velvet. In: Bier C (ed.) Woven from the soul, spun from the heart. Washington DC: Textile Museum, 1987, pp.57–83.

17.

Kodrič Kesovia

Penava

Ž.

Applying image analysis for measuring the density of historical textiles. In: Book of proceedings of the 13th international scientific professional symposium – textile science & economy (ed Petrak S), Zagreb, Croatia, 18 October 2020, pp.265–270. Zagreb: Faculty of Textile Technology, University of Zagreb.

18.

CIETA. Notes on hand-weaving techniques in plain and figured textiles. Lyon: CIETA, 1987, pp.36–37.

19.

CIETA. Vocabulaires techniques textiles. Lyon: CIETA, 1997, p.11.

20.

Becker

Pattern and loom: a practical study of the development of weaving techniques in China, Western Asia and Europe. 2nd ed. Copenhagen: Rhodos International Publishers, 2009, pp.248–261.

21.

Otavský

Wardwell

AE.

Mittelalterliche Textilien II. Zwischen Europa und China. Riggisberg: Abegg-Stiftung, 2011, p.104.

22.

Jemo

Rolich

Application of computer graphics in textile conservation during the documentation process. In: Book of proceedings of the 6th international textile, clothing & design conference – magic world of textiles (ed Dragčević Z), Dubrovnik, Croatia, 7–10 October 2012, pp.632–637. Zagreb: Faculty of Textile Technology, University of Zagreb.

23.

Gill

Images can speak louder than words: communicating conservation effectively. In: Contributions to the Vienna congress. The decorative: conservation and the applied arts, London: International Institute for Conservation (IIC), 10–14 September 2012, pp.114–121.

24.

Peter

Otavský

Salim

, et al. Mittelalterliche Textilien IV. Samte vor 1500. Riggisberg: Abegg-Stiftung, 2019, pp.474–476.

25.

Dilek Himam Er

, et al. Preservation of a textile culture through a digital cultural heritage. Int J Sci Technol Soc 2018; 6: 25–32.

26.

Pagán

Salvatella

MDMG

Pitarch

, et al. From silk to digital technologies: a gateway to new opportunities for creative industries, traditional crafts and designers. The SILKNOW case. Sustainability 2020; 12, 8279.