Seeing Sacred for Centuries: Digitally Modeling Greek Worshipers’ Visualscapes at the Argive Heraion Sanctuary

1.1 Seeing Sanctuaries

“Visualscape” is a catch-all term for what people see, how they see it, and the mental impact of that experience (Llobera 2003). To effectively study a culture’s visualscape, the archaeologist ought first engage with available material, textual, and/or ethnographic evidence and determine how the culture in question saw their surroundings: which viewing experiences mattered, as well as the visual targets. Indeed, our senses are shaped by our sociocultural frameworks. Different smells, sounds, touches, and sights are heightened as a result of our cultural identities, physical abilities, as well as current goals and tasks (Bernardini & Peeples, 2015). Once meaningful viewing experiences, targets, and viewing points are defined, the archaeologist develops an analytical framework. Scholars working on visualscape often use geospatial tools (Bernardini et al., 2013; Landeschi, 2019; Llobera, 2001, 2003, 2004, 2007a,b; Llobera et al., 2010; Richards-Rissetto, 2017; Richards-Rissetto et al., 2023; Susmann, 2019, 2020, 2021). Next, they interpret the results. Archaeologists like Fábrega-Álvarez and Parcero-Oubiña (2019), Fitzjohn (2007), Fredrick and Vennarucci (2021), Given (2004), Rennell (2012), Seamon (2013), Susmann (2020), Van Dyke et al. (2016), and Zhao (2022) discuss how integrated phenomenological observation is a useful way to ground-truth digital results and also imagine the impact of difficult-to-measure factors like vegetation, daylight, and weather. Translating viewing experiences into static pixels is difficult; comparing results across different analyses is recommended (Susmann, 2019, p. 38, 43–44, 156; 2020, pp. 7–9).

To digitally reconstruct the worshipers’ visualscapes at the Argive Heraion, we must begin with identifying culturally relevant viewing experiences. In the context of religion, specific viewing experiences could be quite evocative for worshipers (Neis, 2013, pp. 18–20, 112). Sometimes, invisibility matters. At various points during the Roman Empire, Christians and Jews were forced to meet in secret, and there were no obviously visible signs of practice on the landscape (Deibl, 2020, p. 381). Africans enslaved in America had to conceal their sites of worship in the natural landscape. Their hush harbors formed an invisible sacred landscape amidst one of subjugation (Nunley, 2004, p. 226). Greek worshipers were unconcerned with concealment, at least as far as the exterior sanctuary was concerned. Mystery cults, for example, excluded non-initiates from seeing certain areas of their sanctuary, or even from entering the temenos at all. Locations were not kept secret, and passersby could clearly see sanctuaries like Eleusis.

Just because a sanctuary was visible on the landscape does not necessarily mean it was noticeable. Certain factors made sanctuaries more or less prominent on the landscape. Architecture, for example, played a huge role. A monumental, marble sanctuary like the Argive Heraion would be more noticeable than votives deposited in underground chamber tombs. Prominence is also defined by people. During antiquity, the Greek sanctuaries hosted festival days. Festivals drew huge crowds; noises, smells, and sights emitted from the temenos. Worshipers “activated” the sanctuary’s prominence like the flip of a switch (Ekroth, 2017, p. 33; Stocking, 2017, p. 1, 50–54; Susmann, 2021, p. 24; Williamson, 2014, pp. 181–184).

The natural landscape also shaped viewing experiences, and worshipers chose locations with specific needs in mind (Shepperson, 2017, p. 150). Powerful cities might put sanctuaries on their borders, asking the gods to protect those lines (Cole, 2000, p. 476; de Polignac, 1995, p. 53; Dignas, 2010, p. 164). Cults associated with healing or games could draw large crowds, and sanctuaries might be practically built on easy-to-access, central locations with water sources nearby (Dignas, 2010, p. 164). Sanctuaries could be isolated on islands, meant to provide political asylum or protect the secrets of a mystery cult (Cole, 2015, p. 6; Wescoat, 2012).

Greek cult was rooted in nature, and sanctuaries were often physically linked to specific forests, islands, or water bodies (Bradley, 2000; Buxton, 1992; Larson, 2001; Retallack, 2008). Ancient sources like Pausanias, Xenophon (Cyn.), Apollodorus (Bibl.), and Hesiod (Theog.) reveal why mountains and hills were preeminent in the ancient Greek landscape, in both secular and sacred terms. They were spaces for mining, herding, traveling, and defining boundaries. They were sacred spaces as well. Mountains and hills gave gods easy access to the mortal world; gods transformed earthly mountains and hills into untamed, out-of-control spaces (della Dora, 2008, p. 222; Hom. Il. 1.267-270). Building sanctuaries on the mountains was a regulatory response, allowing the Greeks some assurance for controlling when and where deities might appear (Buxton, 1994, pp. 85–86; Hdt. 6.105; Hes. Theog. 22 ff.; Langdon, 2000, p. 462; Paus. 1.32.2, 2.36.8, 2.43.3; 8.21.4).

For decades, scholars have looked to mountains and hills for physical evidence of these beliefs. Archaeologists working on Crete and the Greek mainland explain how Minoan and Mycenaean worshipers harnessed the mountains and hills, blanketing their sanctuaries with visible symbology. Minoan peak and cave sanctuaries retained lines of sight (Soetens et al., 2001, 2002a,b) and created other-worldly experiences that were distinctive from shrines down below (Briault, 2007; Doxtater, 2009; Karetsou, 1981; Nowicki, 2012; Papalexandrou, 2021; Peatfield, 1983; Soetens, 2009). From inside the temenos, worshipers could not help but notice key landmarks (Psychoyos & Karatzikos, 2015; Susmann, 2019). Scholars like Belis (2015), Cook (1914), Katsarou and Nagel (2021), Nixon (2009), Papalexandrou (2021), Scully (1962), Susmann (2019), and Williamson (2014, 2016) trace these beliefs into the Iron Age. Their research shows how elevated landforms retained a significant role in the cultic landscape, even in situations where a sanctuary had to be placed on flat ground Susmann (2021, pp. 18–20, 27). With this mytho-cultural framework in place, we can infer that seeing a mountain triggered messages about sacredness, just like any man-made temple or altar would (Boutsikas, 2020, pp. 115–116; Darretta & Grubiak, 2022).

Mountains and hills visually signaled palpable and impalpable sacred space (Irvine et al., 2020, p. 109) – but not every mountain, and not all of the time. The physical and mental experience of seeing those hills and mountains – those visualscapes – rested in the hands of Greek worshipers. On the Greek cultic landscape, we can hone into a few key visual experiences that were potentially meaningful to Greek worshipers: noticing sanctuaries from different distances; looking toward the sanctuary as they traveled toward it; and standing in the temenos and seeing the surrounding landscape (Déderix, 2019, p. 201). This list excludes actually witnessing cultic rites from inside the temenos, which is beyond the scope of this article.

2.1 Digital Data Collection

The Argive Heraion’s archaeological record is rich, and we can reconstruct the earliest worshipers in the ninth c BCE, through the Roman adoption of the sanctuary, and beyond. Indeed, the entire Argive Plain has been subject to decades of well-documented surveys and excavations and that matters for a visualscape study. The Argive Heraion’s impact on worshipers depended on those people seeing and experiencing – or at least knowing about – other sanctuaries in the Argive Plain. To detect whether and when worshipers “saw” the Argive Heraion differently over time, we need to compare the Argive Heraion’s visualscape against other cultic features used in the Argive Plain.

I created a GIS dataset that records every single cultic feature in the Argive Plain. The dataset is restricted to non-household contexts, for which there is published material evidence of worship, used between the tenth and second c BCE. These centuries cover the Greek Protogeometric through Hellenistic periods, after Bronze Age Mycenaean culture, and prior to the onset of Roman rule in the Argolid.

The dataset has a wide range of feature types: altars (which could be built of stone or an open-air space for burning and leaving offerings), temples, platforms, etc. A “sanctuary” like the Argive Heraion is composed of several features (altars, temples, etc.), and worshipers would build or dismantle features over time. Splitting sanctuaries into associated features makes it so the geospatial analyses can more precisely account for change over time.^[1]

There were several steps involved: first, retrieving place names from different survey and excavation volumes; next, georeferencing maps or collecting known geospatial coordinates (when available); and then refining these locations using Google Earth. Over the course of two summer seasons, I visited each of these places in person. I recorded my personal experiences walking toward, from, and around the sanctuary space; I recorded photographs and videos that captured views toward and from each feature at varying distances; and I considered how the contemporary vegetation impeded or heightened prominence, as a potential model for what was happening during antiquity.^[2]

2.2 Geospatial Methods

I translated the sociocultural data about Greek sacred visualscapes into contemporary geospatial terms. There are a multitude of geospatial toolkits geared toward visible and physical topography; I settled on three types of analyses: degree of prominence (i.e., noticeability), area of afforded viewshed, and ruggedness of topography (Figure 4). I performed these analyses on every single cultic place in the dataset. If readers would like to replicate any of these analyses, refer to the detailed instructions provided by Susmann (2020) and Susmann (2019, p. Appendix D).

1. Viewshed, or what can be seen from the cultic place (CVSL). I created an observer point file where each place had its own set of observers, strategically situated according to a standardized set of rules. For example, every constructed altar had observers on all four sides, at varying distance intervals to mimic how individuals could stand immediately before, or within the vicinity of an altar. I referenced state plans to avoid placing observers on top of contemporary structures. I measured the visible extent for each observer and next merged these viewsheds together. The resultant raster (i.e., pixel) dataset shows what an individual could see while walking: inside a sanctuary’s temenos, around altars and temples.

2. Visual prominence, based on the number of times a locale was seen across the study area (CVSO). I created a shapefile of 70,378 points, which represents observers moving around the Argolid, spaced 200 m apart. Certain locations were excluded: slopes greater than 49 degrees (the maximum calculated slope for sanctuaries in the Argolid) and confirmed footprints of cultic or non-cultic structures. Using a shared computing cluster, I created viewsheds for each observer point and I merged them into a single raster dataset. This raster records levels of prominence in the Argolid – how many times each location was seen. Using a polygonal boundary shapefile representing a cultic place – for example, the New Hera Temple – I determined how many times it was seen by observers throughout the Argolid. This represents its degree of prominence.

3. Local visual prominence, based on the number of times a locale was seen within 5 km of its boundary (CVSO 5K). This analysis follows the same steps as the CVSO measurement but applies a maximum visible radius of 5,000 m. The resultant cumulative viewshed shows whether each place’s prominence increased or decreased when observers were nearby.

4. Visual prominence, based on how pointy, isolated, and steep the locale was against the horizon (VPO). Using the same observer points from the CVSO/CVSO 5K analysis, I used the ArcGIS’s Skyline tool and modeled what each observer’s visible horizon looked like. Using a Python script published by Bernardini et al. (2013), I compared the rate of change between vertices on these horizon lines and next assigned a comparable score. This score represents that peak’s degree of prominence; if there was no peak, or it was not prominent compared to its neighbors, it received a score of 0. I used these scores to compare which cultic places were located on prominent peaks.

5. Topographic prominence, or how rugged the place was in relation to its surroundings (TP). Ruggedness is based on slope, or how quickly the elevation changes from adjacent locations. I did this calculation using Focal Statistics in ArcGIS. The output calculation is subtracted from the Digital Elevation Model (DEM). The resultant raster describes whether every location in the study area lies above or below their surrounding neighborhood’s average elevation. These values were transferred to cultic places Greeks had used at that location. Topographic prominence does not directly illustrate the viewing experience, but the quantitative data are nonetheless useful for illustrating differences between places.

Figure 4

The Greek visualscape is described based on worshipers’ perspectives looking toward and from the sanctuary, as well as the comparative ruggedness of the physical terrain.

Using JMP statistical software, I normalized the results on a 0–1 scale and created a visualscape model spanning the full chronology (Figure 5), as well as models covering specific periods of interest (Figures 6–8). For any given century, only new or remodeled sanctuaries are recorded so as to highlight how worshipers’ attitudes about the landscape changed, if at all.

Figure 5

Results of the visualscape analyses for the entire Argive Plain, arranged chronologically from the tenth to fourth c BCE. Values are normalized at a scale of 0–1. After the fourth c BCE, no new cultic structures were built within the study area before the end of the Hellenistic period.

Figure 6

Results of the visualscape analyses for the entire Argive Plain, arranged chronologically from the tenth to ninth c BCE. Values are normalized at a scale of 0–1. Only a handful of cultic places are developed in the Argive Plain during this period. They are situated on variable topography and the analyses suggest that these locations were not selected with any particular visual characteristic in mind.

Figure 7

Results of the visualscape analyses for the entire Argive Plain, arranged chronologically from the eighth to seventh c BCE. Values are normalized at a scale of 0–1. During the eighth to seventh c BCE, new cultic places founded on the Argive Plain produced a visualscape foregrounding elevated and noticeable topography.

Figure 8

Results of the visualscape analyses for the entire Argive Plain, arranged from the sixth to fourth c BCE. Values are normalized at a scale of 0–1. By the sixth c BCE, worshipers in the Argive Plain began looking at areas downhill. The surge of cultic places being built in urban areas, such as inside Argos, made the Argive Heraion remarkable in comparison.

The proceeding discussions put the models into context with historical and archaeological data. I use these models to explain how and why worshipers’ visualscapes transformed. I hone into undiscovered nuances and differential experiences: whether the so-called significant views from and toward the Argive Heraion were an integral part of this sacred landscape, what messages were transmitted, and why they changed.

2.3 Methodological Challenges

In a best-case scenario, I would use a 10 m commercial DEM, but this was out of budget. Theoretically, I could have made custom high-resolution DEM for free by flying a drone, but the study area was too large to feasibly cover on my own. I settled on a lower resolution option: a 30 m SRTM DEM and realized that lower resolution imagery comes with benefits. First comes the obvious: it was free. Second, the DEM’s resolution is sufficient enough to capture variable elevations between different cultic places. Third, the lower resolution ensured that these multi-day processes would not fail. The CVSO analysis, for example, involved calculating viewsheds for each of the 70,378 observers in the Argolid. It took 7 days and 6 h to create these viewsheds with Boston University’s Shared Computing Cluster. A higher-resolution DEM would have required significantly more time and CPU.

Deciding how many observer points to use was a similarly difficult decision because I had to weigh increased precision against processing speed and success rates. I tested a range of observer distance intervals in a small test area. Intervals greater than 200 m did not accurately capture changes in terrain; 50 and 100 m intervals crashed. Based on these results, I settled on a 200 m distance interval for the CVSO, CVSO 5K, and VPO tests. Observer points were also placed with worshipers’ behaviors in mind: not overlapping known buildings, water bodies, or slopes greater than 49°. See Susmann (2019, pp. 61–73; 2020, pp. 8–10) for a detailed account of these and other tests.

Depending on where they stood, worshipers’ views could be blocked by buildings, and I grappled with how to model these cases. I created 3D models of every cultic place, using published dimensions (when available), or other known standards. I converted the 3D models into rasters, added them to the DEM with Raster Calculator, and next, ran the CVSO, CVSO 5K, and VPO analyses. Unfortunately, the new DEM was much larger than the original. Even if I set the maximum visible radius to 500 m, it was prone to crashing and could not handle the 200 m interval observer points. Considering my focus on dynamic worshipers, not static points, I reverted back to the original 30 m DEM. In the end, I relied on the 3D models as references for how intra-temenoi structures impacted worshipers’ views. I am eager to revisit the 3D buildings as ESRI continues to develop their Visibility toolkits and Scene Viewer application.

Ideally, this study would integrate vegetation data to examine how seasonal growth impacted visualscape (e.g., whether tree growth impeded worshipers’ views toward the sanctuary). Unfortunately, there are no regional-level vegetation datasets available for the ancient Argolid. Site-level paleoethnobotanical data is also limited. Most these sanctuaries were excavated in the early twentieth century, before paleoethnobotanical sampling was standard practice. Here is where in-person observations become invaluable. We can draw logical connections to the contemporary landscape, hypothesizing about whether and how ancient worshipers’ visualscapes were impacted by vegetation. And so, I visited every cultic place – every site where group worship happened in the Argolid between the Bronze Age through Hellenistic period. I approached the sanctuaries as worshipers would – on foot – and filmed the experience. Hiking up these hills and mountains, crossing these plains, I gained a better understanding of how factors like vegetation, crowds, and altar smoke would have temporarily changed the ancient visualscape (Collar & Eve, 2021; Papalexandrou, 2021, pp. 56–58; Susmann, 2021, p. 13, 23–25, 29; 2022, 2023). I account for these hypothetical scenarios throughout the discussion.

3.1 Where They Worshiped

During the Middle and Late Helladic periods (c. 1900–1100 BCE), Mycenaeans lived on Mount Akraia, in a settlement called Prosymna. Archaeologists found evidence of the settlement in the space of the later monumental sanctuary. They found Early Helladic pottery south of the South Stoa. They found Late Helladic walls in the area between the Old Temple Terrace and North Stoa, between the North and Southwest Stoa, mixed within the foundations of the New Temple of Hera, and in the area between the New Temple of Hera and the West Building (Antonaccio, 1992, p. 89; Blegen & Blegen, 1937, pp. 111–112). A large chamber tomb cemetery accompanied Prosymna, extending about 1 km northeast. Mycenaeans also built a tholos tomb in LH IIIA (c. 1440–1300 BCE), slightly removed from the cemetery (Hope Simpson & Dickinson, 1979, p. 38; Wells, 1990, p. 128) (Figure 9). The Mycenaeans abandoned Prosymna in 1100 BCE when administrative centers in the Argolid and other areas of Greece were dismantled.

Figure 9

Map of Bronze Age through Hellenistic period cultic activity on Mount Akraia.

There is no material evidence for activity on Mount Akraia between the eleventh and tenth c BCE. Granted, this period is defined by widespread political instability and cultural change. Areas like the Argive Plain yield fewer archaeological materials, and this could have happened for a number of reasons: fewer people living in the area and organized into smaller groups, an overall decline in crafting and building, and/or people using easily degradable materials. The age of this excavation is also worth mentioning. Ninety years ago, techniques were comparatively imprecise, and there were fewer scholars publishing about the LBA/EIA transition (Hall, 1995, p. 593; Knodell, 2021, p. 121; Larson, 2016, p. 320; Murray, 2018). In sum, it is entirely feasible that people were using Mount Akraia during the eleventh–tenth c BCE, but their footprints are invisible.

By the ninth c BCE, there is definitive material evidence of Greeks co-mingling with Mount Akraia. They deposited bronze pins inside Prosymna’s historic boundaries (Blegen & Blegen, 1937, pp. 228–229). They built the so-called Old Shrine: an open-air platform approximately 0.1 km southwest of the LH IIIA tholos (Figure 9). The Old Shrine was clearly designed for cultic activity. Archaeologists discovered a 1.2 m area of burning in the center of the platform that was a mixture of bronze and iron fragments, as well as ceramic sherds (Blegen, 1939, p. 411).^[3]

Stylistically speaking, Mount Akraia’s votives differ from contemporary examples in nearby communities (e.g., Argos, Asine, Dendra, Tiryns, or Mycenae) (Hall, 1995, p. 581, 585; Maran & Papadimitriou, 2019, p. 705). In fact, Mount Akraia’s votives resemble examples coming from Aegina, Corinth, Eleusis, and Messinia (Blegen, 1937a, pp. 385–386; Strøm, 2009, p. 90). Still, it is difficult to link Mount Akraia’s earliest worshipers with any specific communit(ies). Votive styles reflect the maker’s origins, less so the worshiper (Morgan, 1993, p. 18). Not every worshiper had to leave a votive, nor would every votive be preserved.

Besides Mount Akraia, there were a handful of cultic sites operating in the Argolid during the tenth–ninth c BCE. Approximately 18 km southeast, worshipers at Kazarma turned a Mycenaean LH I tholos into an altar (Figures 1 and 10). Worshipers also created an open-air sanctuary on the summit of Larissa Hill in Argos (Belis, 2015, Volume 2, p. 116; Courbin, 1966, p. 27, note 23, note 562; Hope Simpson & Dickinson, 1979, p. 51; Roesch, 1953, pp. 99–104, plates 124–128) (Figure 11).

Figure 10

View of the Kazarma tholos, which was reconfigured into a Protogeometric period altar. This photograph was published with permission of the Ephorate of Antiquities of Argolida, Hellenic Ministry of Culture.

Figure 11

On the summit of Larissa is a Venetian fortress; inside this structure, archaeologists found remains of a Protogeometric open-air sanctuary, and following, foundations of a temple dating to the eighth–seventh c BCE. This photograph was taken standing inside the temenos of the Sanctuary of Apollo Pythaeus/Deiradiotes and Athena Polias. This photograph was published with permission of the Ephorate of Antiquities of Argolida, Hellenic Ministry of Culture.

3.2 What They Saw

With the exception of the Kazarma Altar, these emergent cultic places were all visually prominent, albeit in different ways. When prominence is defined based on the landform’s shape against the horizon (i.e., VPO analysis), the open-air sanctuary on Larissa was the only prominent place for cultic activity. This is because of its topography: none of the other tenth–ninth c BCE cultic sites were located on the summit of an isolated and steep hill (Figure 6).

Yet, if prominence is differently defined based on the number of viewing locations from which the place is visible (i.e., CVSO), a different picture emerges. Mount Akraia’s slope is the most prominent, a result of the topography and comparatively larger boundary for worship (Figures 2 and 6). Based on what I have observed with the contemporary vegetation, even thick tree growth was unlikely to have impeded worshipers’ views of the plain below. Worshipers created a sort of “cult stage” (Kerényi, 2015, p. 118) overlooking the Argive Plain; cultural memory and contemporary practice drew their gazes to this particular spot.

4.1 Where They Worshiped

Ongoing building projects at the sanctuary celebrated history and forged physical connections between contemporary worshipers and the mountain’s distant and more recent past. Worshipers built a monumental terrace (i.e., the Old Temple Terrace) in the middle of Prosymna’s boundary, only 0.65 km SE from the ninth c BCE Old Shrine (Antonaccio, 1992, p. 90) (Figures 9 and 12). The terrace’s massive blocks beckon so-called Cyclopean masonry found at Bronze Age citadel sites like Mycenae, Tiryns, and Midea (Antonaccio, 1992, p. 99; Wright, 1982, p. 193) (Figure 13), asserting a visual connection to the Argolid’s Bronze Age past.

Figure 12

The southwestern corner of the Old Temple Terrace, which was photographed from inside the North Stoa. This photograph was published with permission of the Ephorate of Antiquities of Argolida, Hellenic Ministry of Culture.

Figure 13

The cyclopean-style masonry of the Old Temple Terrace. This photograph was published with permission of the Ephorate of Antiquities of Argolida, Hellenic Ministry of Culture.

Interest in the Mycenaean cemetery grew, and worshipers began depositing objects inside the Bronze Age chamber tombs. Objects included bronze figurines, vessels, and a sheathing; votive ceramic cups, jugs, and bowls; a mirror; terracotta spools; and a goat skeleton (Blegen, 1937a,b, 1939; Strøm, 2009, p. 91; Whitley, 2020, p. 27). Sometime in the seventh c BCE, the Old Hera Temple was built on top of the terrace.^[4] The Northern Stoa and Northeast Building were constructed as well (Drerup, 1969, p. 57; Kalpaxis, 1976, p. 46; Mallwitz, 1981, p. 634; Wright, 1982, p. 191).

The diverse collection of votive materials recovered from the Argive Heraion suggests that it was a regional sanctuary, meaning it functioned independently from any of the surrounding settlements, and worshipers were local and non-local (de Polignac, 1995, p. 52; Hollinshead, 2015, p. 44; Knodell, 2021, pp. 212–213; Pedley, 2005, p. 32). Objects came from Corinthia, Laconia, Tegea, and Arcadia: personal ornaments (e.g., pins), vessels, animal figurines, and monumental tripods (Amandry, 1952, pp. 273–274; Strøm, 2009, p. 87).

Consistent with interest in Mount Akraia’s Bronze Age past, most other eighth–seventh c BCE cultic sites in the Argolid were purposefully built over Mycenaean cultic sites and cemeteries. Five Mycenaean tombs have secured Iron Age cultic deposits (Andreadi & Braggiotti, 2003; Blegen & Blegen, 1937; Hägg, 1992; Hope Simpson & Dickinson, 1979) (Figure 1), and seven Mycenaean hill/mountain-top sites were transformed into Greek sanctuaries (Foley, 1988, p. 147; Hall, 1995, p. 597).

Tiryns and Mycenae are citadels (i.e., fortified Bronze Age palaces) located in the middle of the Argive Plain. During the eighth c BCE, both sites transformed into sanctuaries. At Mycenae, eighth c BCE worshipers built a 43 m × 18 m terrace on the highest point of the citadel, and this structure notably overlaps with the megaron (i.e., throne room) (Figures 1, 14, and 15). Like the Old Temple Terrace at the Argive Heraion, worshipers made votive deposits of pottery, bronze pins, and rings and next built a temple on top of the platform (Klein, 2002, p. 105). Nearby, at Tiryns, remains of the Mycenaean megaron were incorporated into an eighth c BCE temple (Figures 16 and 17) (Diniz, 2011, p. 64; Maran & Papadimitriou, 2019, p. 703; Wright, 1982, pp. 191–192, 195, Foley, 1989, #2986).

Figure 14

Mycenae is nestled in between two hills: Psara and Protifis Ilias. This photograph was published with permission of the Ephorate of Antiquities of Argolida, Hellenic Ministry of Culture.

Figure 15

Remains of Mycenae’s Hera Temple foundations, looking from the southeast. This photograph was published with permission of the Ephorate of Antiquities of Argolida, Hellenic Ministry of Culture.

Figure 16

The citadel of Tiryns, looking north. This photograph was published with permission of the Ephorate of Antiquities of Argolida, Hellenic Ministry of Culture.

Figure 17

Area of the megaron at Tiryns, where a seventh c BCE temple was constructed. This photograph was published with permission of the Ephorate of Antiquities of Argolida, Hellenic Ministry of Culture.

Mount Arachnaion demarcates the southeastern boundaries of the Argive Plain (Psychoyos & Karatzikos, 2015, p. 261) (Figures 1 and 18). The saddle-shaped summit has twin peaks. The highest peak, Profitis Ilias, was once the site of a popular Mycenaean peak sanctuary. In the early eighth c BCE, worshipers transformed this space into an open-air altar (Foley, 1988, p. 150; Psychoyos & Karatzikos, 2015, p. 261) (Figure 19).

Figure 18

View of the modern city of Argos, photographed from the summit of Larissa Hill, looking northeast. Mount Akraia is the central mountain on the horizon. On a clear day, the Argive Heraion is visible. This photograph was published with permission of the Ephorate of Antiquities of Argolida, Hellenic Ministry of Culture.

Figure 19

The summit of Mount Arachnaion, within the boundaries of the peak sanctuary. The walled structure in the background is a contemporary chapel. This photograph was published with permission of the Ephorate of Antiquities of Argolida, Hellenic Ministry of Culture.

Moving 9 km southwest, there is a much smaller hill called Profitis Ilias Katsingri (Figure 1). During the Bronze Age, Mycenaeans used the pointy, rocky summit for an open-air sanctuary (Figures 20 and 21). In the seventh c BCE, Greeks came to the same summit and created their own open-air sanctuary. They burned fires, made sacrifices, and deposited bronze figurines and pottery sherds. At some point during the Archaic period (c. 700–480 BCE), worshipers decided to add a 6.6 m × 13 m temple. The foundations are currently covered by a modern Orthodox chapel (Foley, 1988, p. 150; Hall, 1995, p. 597).

Figure 20

Profitis Ilias Katsingri’s rocky summit. This photograph was published with permission of the Ephorate of Antiquities of Argolida, Hellenic Ministry of Culture.

Figure 21

On Profitis Ilias Katsingri, there are unexcavated remains of a temple, partially visible under the foundations of a modern chapel. This photograph was published with permission of the Ephorate of Antiquities of Argolida, Hellenic Ministry of Culture.

The last two places are located on hills in Argos: the Aspis and Larissa. These hills overlooked ancient Argos (now a major city), the Argive Plain, and beyond that, Mount Akraia. Mycenaeans had a settlement and cemetery on the Aspis. In the eighth c BCE, worshipers built the Sanctuary of Apollo Pythaeus/Deiradiotes and Athena Polias (Papahatzis, 1978, p. 106) (Figures 1 and 11). Larissa is unique from the rest of these sanctuaries because there is no evidence of Mycenaean worship, but for the eighth–seventh c BCE Greeks, it was still a relic. The earliest evidence of worship at Larissa dates to the tenth–ninth c BCE, when Greeks worshiped in the open air on the summit. In the seventh c BCE, Greeks dedicated temples to Zeus and Hera on the summit (Vollgraff, 1928, p. 318, plate VII, plate VIII). Even without known ties to the Bronze Age, Greeks’ attention to this historic landmark mirrors what happened on the Aspis, Mount Akraia, or any of the other eighth–seventh c BCE sanctuaries. As time goes on, definitions of “old” will inevitably extend, and Larissa demonstrates how Argive notions of ancestry were fluid and ever-changing.

4.2 What They Saw

Measuring this landscape with CVSO and VPO analyses reminds us about the personal nature of vision that different people saw landforms in different ways. According to the CVSO analysis (i.e., the place’s prominence based on the number of viewing points), the vast majority of these sanctuaries were noticeable from at least 25% of the entire Argolid region (Figure 7). In other words, they were easy to spot from a distance. VPO (i.e., prominence based on the location’s shape against the horizon) reminds us that some of this group are further distinctive. Mount Arachnaion, Profitis Ilias Katsingri, and Larissa are uniquely shaped in comparison to their surroundings.

The TPL (i.e., degree of ruggedness) and CVSL (i.e., visible extent) results, as well as field observations, demonstrate the link between looking toward a sanctuary and looking from it – prominent sanctuaries provide extensive views of the surrounding landscape. At some sanctuaries, like Mount Akraia, worshipers could clearly see other sacred sites and settlements in the region – provided they knew where to look. In other words, worshipers were not practicing in isolated sanctuaries across a broad and disconnected area; these were nodes from which worshipers could visualize a network of heritage and identity.

Putting these data into context with cultural practices, we understand that the Greek worshipers’ sacred landmarks could be amplified by cultural memory. Worshipers were looking toward Mount Akraia and the Argive Heraion because they knew about it. Cultural knowledge about this site – all tied into their reverence for the past – bolstered the already-prominent summit. Framed within the entire Argive the sacred landscape, however, the Argive Heraion was not an outstanding landmark – at least not during these formative centuries. Cultural memory was tied into these mountains, amplifying sacred summits moreso than the rest. The temples and altars were monumental, human-made markers used to demarcate where worship happened, but the Greeks recognized the mountains first. Mount Akraia – or Profitis Ilias Katsingri, or the Aspis, or Larissa – were discernible landmarks, signaling the finish line toward the sanctuary. Monumentality was, now, tied into the sacred landscape and used to demarcate specific parts of the mountain for worship. When worshipers saw a temple or altar, they understood what took place there.

5.1 Where They Worshiped

By the sixth c BCE, Argos had transformed into a powerful city-state and the Argive Heraion was a landmark of its power. As a territorial sanctuary, the Argive Heraion protected Argos’ vast territory, quite literally looking over the towns and cities they had conquered (Amandry, 1952, pp. 273–274; Maran & Papadimitriou, 2019, p. 710; Strøm, 2009, p. 87).

Using prowess, money, and creativity, Argos transformed the Argive Heraion’s physical and metaphysical footprint (Hall, 2002, pp. 103–104; Maran & Papadimitriou, 2019, p. 710). Argos wrote themselves into the sanctuary’s history: ever-powerful, worshiping at the Argive Heraion from the outset.

The myth of Cleobis and Biton, for example, describes two pious brothers who help their mother reach the sanctuary by dragging her across the Argive Plain in an oxen cart:

[…] there was a festival of Hera in Argos, and their mother absolutely had to be conveyed to the temple by a team of oxen. But their oxen had not come back from the fields in time, so the youths took the yoke upon their own shoulders…they drew the wagon, with their mother riding atop it, traveling five miles until they reached the temple. (Hdt. 1.31.2)

This story is a crafty piece of fifth c BCE Argive propaganda (Maran & Papadimitriou, 2019, p. 711). Cleobis and Biton represent Argos: the birthplace of tenacious souls with outstanding physical and moral character. Beginning at Argos, they plowed through the plain; they watered the land with their sweat and tears. Yoked together, Cleobis and Biton demarcated the extent of Argive territory and influence. They tied the city and sanctuary in turn. Stories like these helped redefine visualscape; the Argive Heraion was signaling new visual messages to worshipers.

When in 494 BCE Cleomenes, King of Sparta, led an attack on Argos, he “was taking omens in Hera’s temple [at the Argive Heraion] a flame of fire had shone forth from the breast of the image, and so he learned the truth of the matter, that he would not take Argos (Hdt. 6.82).” In the 460s BCE, Mycenae and Argos fought to control the sanctuary. The Mycenaeans “would not be subservient to the Argives as the other cities of Argolid were…and they kept disputing with them also over the shrine of Hera (Diod. Sic. 11.65).” The narratives surpassed the citystate’s lifespan. The Roman geographer Pausanias tells us: “The oldest image [in the Argive Heraion] is made of wild-pear wood, and was dedicated in Tiryns by Peirasus, son of Argus, and when the Argives destroyed Tiryns they carried it away to the Heraeum (Paus. 2.17.5).”

Looking at the material record, the most notable change happened in 423 BCE, when the Old Temple of Hera accidentally burned down. It was immediately replaced with the so-called New Temple of Hera (Maran & Papadimitriou, 2019, p. 710) (Figures 3 and 9). The New Temple of Hera was notably set off its predecessor’s footprint, slightly downhill. There are many possible motivations for choosing a new location: concerns about structural stability after the fire, or perhaps a desire to increase accessibility with a shorter, less steep walk. Practicality aside, the New Temple is another example of how Greeks living in the Argolid were so attached to their region’s collective memory (Pfaff, 2003, p. 6). Adjacency mattered, and so did protecting the burned remains. Even in their ruined state, the Greeks saw the Old Temple and Terrace as an epoch that must be preserved and revered.

Across the plain at Argos, worshipers built three new cult places, all situated on flat ground, wrapping around the base of Larissa (i.e., the Temple and Altar of Wolf Apollo and the Altar of Aphrodite). According to Maran and Papadimitriou (2019, p. 709), the Temple of Wolf Apollo was purposefully located alongside one of Argos’ main exit routes, notably adjacent to a wealthy Late Geometric period cemetery. Yet again, worshipers were forging spatial ties with preceding generations, albeit fixated on low-lying, easy-to-access sanctuaries inside of the city.

Across the rest of the plain, all of the historic sanctuaries were either abandoned or destroyed. Mount Arachnaion’s votive deposits markedly declined in the sixth c BCE; the sanctuary was used sporadically through the Roman period (Foley, 1988, p. 150). Profitis Ilias Katsingri’s sanctuary was gradually abandoned, with no votive deposits made after the Archaic period (i.e., 480 BCE) (Kilian, 1990, p. 190). Tiryns and Mycenae were sacked by Argos during the fifth c BCE and its temples were destroyed (Hall, 1995, p. 597, 608; Papahatzis, 1978, p. 115). Sometime during the third c BCE, Mycenae’s Hera temple was rebuilt. The effort to destroy and then repair Mycenae was a clear assertion of Argos’ everlasting dominance (Klein, 2002, p. 104).

Before the fifth c BCE, a Sacred Way connected Mycenae’s Hera Temple, Chaos Shrine, and Bronze Age tholoi with the Argive Heraion. Once Mycenae’s Hera Temple was destroyed in the fifth c BCE, the route became obsolete, at least in terms of cult. There was a new pathway for pilgrimage, and it traces Kleobis and Biton’s steps: worshipers exited Argos, crossed the Argive Plain, and climbed up Mount Akraia to the Argive Heraion (Hall, 1995, p. 612) (Figure 22). Once there, worshipers looked out on a new iteration of the Argive Plain, and Argos was at the other end. All of these changes reflect Argos’ strategy to weave new landmarks and pathways into their territory, to control narratives about their roles in the region’s history.

Figure 22

The area surrounding Mycenae, including the Enyalios Sanctuary and the Chaos Shrine. M3 is a Mycenaean route adopted by the Greeks.

5.2 What They Saw

Previous scholars have linked the Argive Heraion’s visual prominence to its political prowess (de Polignac, 1995, p. 33; Foley, 1988, p. 154; Kelly, 1976, pp. 62–63; Wright, 1982, p. 199). For example:

Situated midway between Argos and Mycenae on a low hill, this sanctuary commanded a good view of the surrounding plain and, while several other sanctuaries of Hera existed in the Argolid, none had the prominence of the Heraion. (Foley, 1988, p. 135)

These studies predate Hall (1995) and Strøm (2009), whose careful analyses explain how the earliest worshipers were not from Argos. These earlier studies incorrectly characterize Argos as the sanctuary’s founders. In other words, their characterization of a politically charged visualscape is correct – but only after the sixth c BCE, when Argos did eventually take control.

Inserted into this wider spatial and chronological framework, we realize that the Argive Heraion’s visual prominence was extraordinary in the Argive Plain – but not before the sixth c BCE (Figures 5 and 8). This change happened as a result of other sanctuaries. Prior to the sixth c BCE, worship in the Argolid tended to happen on isolated, elevated landforms. Prominence – whether defined by CVSO or VPO approaches – was a key feature of these sanctuaries, and this is why the vast majority have prominence scores above 0. Prominence, and the inevitably vast viewsheds it produced, were the norm.

After the sixth c BCE, the cultic landscape changes, and it is apparent in the archaeological and historical records, as well as the visual prominence analyses (Figure 8). The destroyed or abandoned sanctuaries like Tiryns, Mycenae, and Arachnaion were not rendered invisible; the places still existed in the collective memory, but worshipers inevitably saw them in a different light. They were relics of the past, removed from this generation’s sacred landscape.

The Argive Heraion was part of the historic landscape, but it was part of the contemporary sacred landscape as well. In the sixth c BCE, we see a new version of the Argive Heraion, and it changed because of the changes happening around it. The Argive Heraion contrasted with every other sanctuary in the Argive Plain. It was the largest, most elevated, and provided worshipers with an undeniably impressive view of all the territory Argos had conquered. Worshipers could see the Argive Heraion from Argos; they could see Argos from the Argive Heraion.

The juxtaposition is clear when we look inside Argos’ boundaries, where new, low-lying sanctuaries were built in the middle of the bustling urban fabric. At these places, worshipers had an entirely different viewing experience. In contemporary Argos, homes and businesses block views of the ancient sanctuaries, and we can imagine similar circumstances during antiquity. Worshipers would only be able to see the sanctuaries if they were nearby. For anyone outside of the city, seeing the sanctuaries from any distance would be impossible. They would rely on other landmarks and visual cues: smoke emanating from the altar, Larissa, and the city itself.